From burek021 at gmail.com Tue Dec 1 02:05:02 2015 From: burek021 at gmail.com (burek) Date: Tue, 1 Dec 2015 02:05:02 +0100 (CET) Subject: [Ffmpeg-devel-irc] ffmpeg-devel.log.20151130 Message-ID: <20151201010503.275D42AD601C@apolo.teamnet.rs> [04:05:40 CET] wtf is this http://www.acm.org/about/se-code [09:29:29 CET] wm4: https://trac.ffmpeg.org/ticket/5047 <- that's for you I think :-) [09:29:50 CET] wm4: cause when looking through the commits i saw some by you in that direction [09:32:43 CET] fritsch: that's not even a mp3, that's a mp4, so my dealings with the mp3 demuxer can't have broken it [09:33:07 CET] it's probably a problem with enabling parsing [09:34:11 CET] also getting 403 on the sample url [09:35:04 CET] wm4: retry fixed permission [09:35:06 CET] sorry for that [09:35:34 CET] plays nicely in the browser even :-) [09:39:33 CET] and yes it's just mp3 in mp4 [09:48:05 CET] unconditionally enabling full parsimg helps [09:48:32 CET] supposedly that breaks other things [09:50:15 CET] personally I would enable it and then try to fix the parser to not break things, even better if someone has a sample handy of such a case right now [09:58:13 CET] nevcairiel: I'd feel better about that than about the arbitrary and unreliable heuristic that enables parsing [10:14:31 CET] https://github.com/rcombs/FFmpeg/commits/auto-bsf now with mpegts support [10:15:01 CET] also, I changed mpegts's write_header to init because it never actually writes anything, which also lets me remove some duplicated cleanup code :D [10:17:13 CET] so that leaves AVI and WTF still using ff_check_h264_startcode and needing auto-filtering for AVC, and MOV and FLV still needing auto-filtering for AAC [10:38:10 CET] so what happened to the idea of dropping -fno-tree-vectorize? [11:25:04 CET] nevcairiel: I'm good at typing; meant ff_stream_& [11:25:11 CET] i figured [11:25:24 CET] so where does one get truehd specs? [11:25:38 CET] unless you have a better idea for the name [11:25:41 CET] one doesnt [11:25:47 CET] `ff_stream_&` feels super awkward [11:26:00 CET] specs? [11:27:27 CET] but I guess it just feels that way because `av_stream_&` mirrors the struct name [11:27:31 CET] so w/e [11:28:16 CET] nevcairiel: why not? [11:28:20 CET] is it like the DVD spec? [11:28:25 CET] wm4: from kodi's bitstream parser code [11:28:34 CET] dunno, but it seems rather carefully guarded [11:29:12 CET] wm4: https://github.com/FernetMenta/xbmc/blob/master/xbmc/cores/AudioEngine/Utils/AEBitstreamPacker.cpp [11:29:13 CET] whats so special about the kodi code thats not in the ffmpeg decoder? :d [11:29:32 CET] yeah ffmpeg has such code too [11:29:45 CET] nevcairiel: it was there before and no one yet replaced it [11:29:47 CET] https://github.com/FFmpeg/FFmpeg/blob/master/libavformat/spdifenc.c [11:30:01 CET] when I worked on Atmos extensions last year, I googled for alternative truehd implementations [11:30:07 CET] all of them were based on the ffmpeg code [11:30:18 CET] see the author of the ffmpeg code [11:30:24 CET] he is also kodi audio dev [11:32:02 CET] so in short, all that is known about truehd/mlp can be found in ffmpeg, to get some extra info i had to get the disasm on [11:34:15 CET] nevcairiel: the reason we still have our part of the code is the architecture ... which needs some work to get the ffmpeg bits into place [11:34:40 CET] isnt that always the reason =p [11:34:56 CET] I don't understand that... the ffmpeg code could just be used in place there [11:35:17 CET] spdifenc is a big hack API-wise (it's a muxer, really?) [11:35:21 CET] i've been using spdifenc from avformat for a couple years now, and can confirm it works for all dolby and dts codecs, i didnt have hardware to test aac [11:37:43 CET] Anssi: ^^ our reason to not use it was "lack of time right"? I cannot remember the architecture argument [11:41:29 CET] fritsch: unfortunately I don't remember why gnif reimplemented it in our own code... I remember the ffmpeg code working just fine, but maybe I just forgot [11:42:55 CET] the code has grown some extra E-AC-3 support now, I think, though, that ffmpeg is missing, but it is somewhere in my long todo list to port it to ffmpeg [11:47:01 CET] ffmpeg supports eac3 in spdifenc [11:47:10 CET] my software uses it fine [11:47:27 CET] yeah - then lack of time is left as argument :-) [11:48:05 CET] although i'm not 100% sure if it works perfectly, as a few select people reported that eac3 didnt work for them, but it works on my hardware .. maybe their hardware is just more picky, or just plain out doesnt support it, hard to judge [11:48:51 CET] yeah - ours also works - just revisited it with pulseaudio some days ago as we restructed stream_type handling [11:49:00 CET] cause of android wants to have non IEC streams [11:50:32 CET] I never can get pulse to pass through compressed audio [11:50:43 CET] wm4: then check our AESinkPULSE [11:50:44 CET] :-) [11:50:52 CET] that's not the problem [11:50:59 CET] and follow the wiki: set device to stereo and enable dts/ac3/eac3 [11:51:03 CET] the problem is configuring pulse in a way it does it at all [11:51:08 CET] ah [11:51:12 CET] yeah i have screenshots in the wiki [11:51:16 CET] lol [11:51:25 CET] and the sink which works :-) [11:51:27 CET] so what is left? [11:51:27 CET] so you're confirming my point or what? [11:51:45 CET] (that it's a pain to setup pulse to do that if you need screenshots) [11:51:46 CET] just said: we are succesfully using PA for DTS /AC3 / EAC3 [11:51:48 CET] users need to do careful setup of audio settings for bitstreaming to work? yeah comfortable much :) [11:52:06 CET] wm4: open pavucontrol, check your hdmi device click ac3, dts, eac3, make sure it is stereo profile [11:52:16 CET] http://kodi.wiki/view/PulseAudio#Passthrough_Mode [11:52:33 CET] note that HD audio doesnt use stereo, it uses more channels, but maybe you just dont do HD [11:52:40 CET] hd audio won't work with PA [11:52:48 CET] "REMEMBER: To use passthrough you need to be using 2.0 channel configuration via pavucontrol (e.g. Digital Stereo (HDMI)" [11:52:49 CET] lol [11:52:52 CET] that's why I talk about dts, ac3, eac3 [11:53:08 CET] what does hd audio mean? [11:53:13 CET] dts-hd, truehd [11:53:15 CET] truehd and dts-hd [11:53:19 CET] everything that needs more than 2 pcm channels [11:53:22 CET] to be sumbitted [11:53:28 CET] lol [11:53:30 CET] in our case 8 + 192 khz [11:53:50 CET] pulse, how to make your audio not work [11:54:27 CET] wm4: anything left? :-) the rest is in the AESinkPULSE [11:54:49 CET] I implemented this years ago [11:54:56 CET] yeah, me too [11:55:00 CET] but every time I test I have to struggle with pulse [11:55:05 CET] hehe [11:55:14 CET] it's straight forward [11:55:17 CET] no [11:55:18 CET] only for eac3 you need to take care [11:55:29 CET] with old api using 4*encoded_sample_rate [11:55:34 CET] the new one does it automatically [11:55:55 CET] what old and new API? [11:55:57 CET] I always have to suppress an evil laughter if i have to hear about the "fun" with linux audio [11:56:13 CET] yeah - not to mention the fun of windows directsound api ... [11:56:16 CET] nevcairiel: I have my own share of fun with other platforms too [11:56:27 CET] which supports also only dts, ac3 and no eac3 [11:56:33 CET] and wasapi is a big mess [11:56:45 CET] if you need to handcode the delay measurements ... [11:56:45 CET] i can bitstream even HD through a directsound renderer [11:56:54 CET] nice - we can't [11:57:24 CET] https://github.com/FernetMenta/xbmc/blob/master/xbmc/cores/AudioEngine/Sinks/AESinkPULSE.cpp#L627 <- wm4 [11:57:37 CET] prior to 2.0 [11:58:57 CET] nevcairiel: https://msdn.microsoft.com/en-us/library/windows/hardware/ff538534(v=vs.85).aspx <- we use this [11:59:06 CET] ac3 for both so that directsound "eats" it [11:59:12 CET] and then the header decides [11:59:18 CET] fritsch: so you just hope PA assumptions somehow matches kodi assumptions about the raw format? [11:59:28 CET] i would have to find the code that deals with this [11:59:29 CET] nope - i know about the assumptions [11:59:32 CET] cause I read the PA code [11:59:47 CET] wm4: which assumptions btw.? [12:00:28 CET] it's not that complicated ... it takes eac3 format type iec encapsulated, stereo speaker config + matching format [12:00:31 CET] that's it [12:00:33 CET] and obviously the samplerate [12:01:05 CET] ah ok, the kodi params are set in info? [12:01:41 CET] yes [12:02:06 CET] encoding, sample_format, channel_map, samplerate [12:02:33 CET] that's it the rest is "buffer" management and api management [12:02:34 CET] that's weird, my code doesn't even use pa_sample_spec [12:02:45 CET] (at init anyway) [12:02:53 CET] you most likely don't use pa_stream_new_extended [12:02:54 CET] then [12:03:11 CET] I do [12:03:17 CET] argh wait [12:03:21 CET] see where we use it [12:03:21 CET] it takes a pa_format_info [12:03:26 CET] ours, too [12:03:31 CET] https://github.com/FernetMenta/xbmc/blob/master/xbmc/cores/AudioEngine/Sinks/AESinkPULSE.cpp#L634 [12:03:34 CET] just used here [12:03:35 CET] for validation [12:03:48 CET] seems pointless [12:03:53 CET] and for estimating bytespersecond, framesize and so on [12:03:59 CET] not really [12:04:09 CET] that way you can make sure that PA likes what you setup [12:04:17 CET] I have my own functions for calculating bitrate etc. [12:04:18 CET] and can easily get the management data you need [12:04:28 CET] yeah - if you like implementing it yourself [12:04:38 CET] i thought letting the API do as much as possible [12:04:43 CET] give me less code to maintain [12:05:10 CET] mpv's pulse sink source file is slightly shorter than kodi's [12:05:15 CET] i think so [12:05:22 CET] we have a lot management stuff in there [12:05:24 CET] device hotplug [12:05:26 CET] enumeration [12:05:28 CET] me too [12:05:36 CET] the addpackets method should be the same size [12:05:55 CET] well, not that it means anything, but I still don't think the sample_spec thing is really needed [12:05:58 CET] and some AE_TYPE to pa_type static conversion stuff could be moved into a util method [12:06:17 CET] http://maemo.org/api_refs/5.0/5.0-final/pulseaudio/sample_8h.html#0d9ad972ee71ef6bff8aaafd44fdb229 [12:06:20 CET] yeah [12:06:30 CET] most likely you would fail when you try to create the stream [12:06:47 CET] I would hope so [12:07:51 CET] the site you linked doesn't have pa_stream_new_extended [12:09:07 CET] it was about the valid [12:09:10 CET] http://freedesktop.org/software/pulseaudio/doxygen/stream_8h.html#acf41f201c5e6ac94b7fcbc8626b6c933 [12:09:14 CET] "While connecting, the server will select the most appropriate format which the client must then provide." [12:09:16 CET] whatever that means [12:09:27 CET] (one format from the list, or possibly something very different?) [12:09:36 CET] we don't want something different [12:09:45 CET] and therefore ask before [12:09:48 CET] if the combination works [12:09:51 CET] where? [12:09:58 CET] the _valid_ call [12:10:09 CET] but that's probably client-side only, without involving the server [12:10:28 CET] yeah - but the server side is covered with the enumeration [12:10:35 CET] there we know what we can provide / server has [12:10:49 CET] -> lunch time :-) [12:10:56 CET] so the very moment the spec is valid [12:11:04 CET] and we have setup a format our sink can do (enumeration) [12:11:08 CET] -> it should work (TM) [12:11:15 CET] why not use this? http://freedesktop.org/software/pulseaudio/doxygen/format_8h.html#a9efd301cfd24b37737c6a2e2b47c9261 [12:11:31 CET] fritsch: well it could change after enumeration [12:31:34 CET] ffmpeg 03Paul B Mahol 07master:376a3bdb7bbe: avfilter/af_sidechaincompress: add level_in and level_sc options [12:31:34 CET] ffmpeg 03Paul B Mahol 07master:6b978dadc654: avfilter/af_sidechaincompress: do not require writable input frames for acompressor filter [12:55:18 CET] wm4: see the our registering for the callback [12:55:23 CET] wm4: we get informed when this happens [12:55:26 CET] and change accordingly [12:55:44 CET] and ask AE factory to care for it, e.g. reinitializing us [13:59:18 CET] nevcairiel: yeah the ffmpeg e-ac-3 code works fine for normal streams, but it doesn't have any substream handling code, so e-ac-3 streams using substreams with independent streams might not (I haven't looked into this yet) [14:00:40 CET] i guess they might not work, i'm not sure, luckily discs using that concept are exceedingly rare [14:00:55 CET] yeah, substreams don't work at all [14:02:44 CET] adding support for dependent substreams (e.g. 7.1 eac3) will require some internal tweaks to how channels are handled, but it shouldn't be too bad [14:02:55 CET] for multiple independent substreams& I'm not even sure how that should work [14:03:01 CET] we're talking about bitstreaming [14:03:33 CET] oh, like spdif passthrough? [14:03:39 CET] yes [14:04:21 CET] I wonder how multiple independent substreams should be treated [14:04:22 CET] you'd probably want those as separate AVStreams, really [14:04:48 CET] but there's no mechanism for a parser or a decoder to output multiple streams [14:05:06 CET] there probably should be, though [14:05:12 CET] also for things like H.264 with embedded 708 captions [14:07:43 CET] here be all sorts of dragons [14:33:33 CET] why is fate-api-seek depending on fate-lavf? [14:35:54 CET] mp3 in mp4 is a hack... [14:37:50 CET] Compn: in general or in ffmpeg realization? [14:38:23 CET] everything in multimedia is a hack [14:38:27 CET] lol [14:38:27 CET] especially Compn's existance [14:38:42 CET] concerning "the standard" it is perfectly fine [14:38:48 CET] how is it a hack [14:38:51 CET] doesnt it have a spec even [14:38:57 CET] it has - therefore I asked [14:40:10 CET] anyway my funny little patch breaks the sample that was pointed out [14:40:36 CET] and I'm too lazy (which actually means: busy) to analyze and fix it [14:40:57 CET] as 2.7 worked - a bisect should find the bugger [14:41:02 CET] i can do that if it's welcome tonight [14:41:45 CET] might be a patch-that-fixes-one-sample-breaks-another thing [14:42:05 CET] funny the linked patch also talks about handbrake encoder [14:42:11 CET] the linked bug [14:42:34 CET] it's probably the third bug I file that happens with users using handbrake [14:42:53 CET] oh yes we discussed it with a handbrake dev too [14:42:55 CET] which I think also uses ffmpeg under the hood? [14:43:15 CET] so that might be an ffmped 2.7 produced "stream" cannot be decoded properly with 2.8 [14:43:18 CET] that would suck [14:43:35 CET] handbrake likes to play a bit loose with the spec sometimes, the argument pretty much ended in "but the spec says this is valid" =p [14:44:02 CET] for us we always get: does not play in kodi 16 - but worked in kodi 15 [14:44:08 CET] fix it, it s regression they say :-9 [14:46:11 CET] does quicktime play it ? [14:46:56 CET] without audio my colleague said [14:47:00 CET] i don't have quicktime [14:47:21 CET] if quicktime cant play it, then its not valid mp4 :P [14:47:25 CET] mpv based on ffmpeg 2.7.x works [14:47:31 CET] ffplay 2.7.x also works [14:47:41 CET] i realize that we can play such files of course [14:47:45 CET] and that it is a regression [14:47:49 CET] just sayin... [14:47:52 CET] the biggest problem here is: [14:47:57 CET] handbrake produces such stuff _right now_ [14:47:59 CET] I bet mplayer with its old builtin mov demuxer can play it [14:48:07 CET] of course mplayer can still play it :P [14:48:09 CET] I did not try [14:48:25 CET] i use kodi - if it fails I try mpv - if that also fails -> file is broken [14:48:27 CET] :-) [14:48:27 CET] because mplayer uses mpg123, so mpg123 somehow pieces the data together [14:48:28 CET] dont we have a handbrake author here on irc ? [14:48:49 CET] fritsch : we should strongarm this person into disabling mp3 audio [14:49:43 CET] jstebbins seems the main author [14:49:45 CET] cannot find him here [14:49:58 CET] fritsch, he is j45 [14:50:06 CET] j45: ping [14:50:16 CET] Daemon404: thx - yeah yours pinged him already [14:50:28 CET] Action: Compn goes afk [14:50:36 CET] Compn: haha [14:50:45 CET] gotta run, 8:50 am on monday. [15:01:57 CET] just tried it with mplayer, fails on at least 1 of the samples too [15:06:07 CET] kierank: whats status? [15:06:33 CET] durandal_1707: ? [15:06:51 CET] of vc5 [15:07:39 CET] dunno it's a weekend project [15:07:55 CET] smpte people are investigating whether I can get reference decoder [15:09:49 CET] they contacted you? [15:32:29 CET] Yes I asked for samples [15:33:07 CET] Basically I understand bitstream now [15:33:17 CET] Haven't found out how they choose Huffman tables [15:33:20 CET] wait, do you mean the reference encoder source, or just a binary? [15:33:45 CET] Reference Encoder source for vc-5 and samples (not the same as cfhd) [15:36:16 CET] vc-5 is just cfhd with extra bytes? [15:37:04 CET] cfhd is in avi, and vc-5 is raw format? [15:37:34 CET] Vc-5 is a different format inspired by cfhd [15:37:51 CET] But they share a few features [15:38:33 CET] I will upload my tree to github later [15:39:10 CET] I basically have a bitstream parser and a semi-working lowpass decoder [15:40:18 CET] for cfhd samples we have? [15:42:22 CET] I have 3 [18:09:21 CET] so many mjpeg variants [19:54:34 CET] if anyone cares, vs2015 update 1 final is available now [19:57:39 CET] are there release notes or anything somewhere? interested about the supposed clang stuff [19:58:09 CET] thats not part of the update, but an extra addon, didnt find that one yet [19:58:19 CET] there will probably be a blog post up today at http://blogs.msdn.com/b/vcblog/ [19:58:25 CET] with more useful information than the marketing stuff [19:58:57 CET] another interesting feature is that VS can now talk to GDB and play debugging frontend with GDB as a backend [19:59:13 CET] breakpoints, memory inpsection, et al [20:00:10 CET] also, there is a tool now to only install the compiler and sdk without the IDE [20:00:22 CET] which might be useful for build systems or wine setups [20:10:32 CET] nevcairiel, im one of those freaks that only uses the ide for debugging [20:10:40 CET] and cl.exe directly [20:12:29 CET] my fate box would've probably used such a tool [20:12:33 CET] but alas its already setup now [20:13:01 CET] should make it easier to update to newer versions, though [20:13:19 CET] ... can you actually uninstall now? [20:23:29 CET] the toolchain-only version should be uninstallable [20:23:43 CET] since it doesn't have all the extra crap [20:23:55 CET] hmmm [20:24:12 CET] and they're switching to sqlite for intellisense etc. [20:24:21 CET] that reminds me to t urn on that option [20:24:24 CET] its apparently much faster [20:24:24 CET] so maybe one day in the future there won't be a million sql server things that have to be uninstalled [20:26:00 CET] Doesn't it use an embedded sql express server, unless you explicitly enable a full one? [20:26:10 CET] it does [20:26:21 CET] but its still installed as a separate thing [20:26:22 CET] So basicaly MS-SQLite [20:26:30 CET] and apparently SQLite is faster [20:26:43 CET] so they added an option to switch to SQLite in update 1 [20:26:50 CET] its still considered experimental, so no default [20:26:54 CET] but in the future it may move there [20:43:02 CET] ..installing the upgrade takes as long as ever [20:47:45 CET] speaking of compilers, gcc 6 reached stage 3 (feature freeze) and fate passes [20:48:35 CET] 5.3 is also scheduled for release this or next week [20:48:38 CET] as is binutils 2.26 [20:51:38 CET] do they specifically include fate in their tests now, or why is it passing even though it's not even released yet? [20:52:31 CET] a bunch of people here test fate on gcc master builds and instantly report breakage, that usually helps [20:54:48 CET] there fate box updated, lets see what happens [21:01:56 CET] i'm running a gcc trunk fate box, yeah [22:42:20 CET] Is it ok to use complex.h? [22:53:13 CET] probably not [22:53:29 CET] if you do, add a configure test and dependency and everything [00:00:00 CET] --- Tue Dec 1 2015 From burek021 at gmail.com Tue Dec 1 02:05:01 2015 From: burek021 at gmail.com (burek) Date: Tue, 1 Dec 2015 02:05:01 +0100 (CET) Subject: [Ffmpeg-devel-irc] ffmpeg.log.20151130 Message-ID: <20151201010501.D9E4A18A00F2@apolo.teamnet.rs> [04:42:29 CET] does -c copy loosy bitrate on videos? [04:46:00 CET] hey, anyone know why when i start a stream it starts right but over time the audio/ video starts to drift out of sync ? [04:46:04 CET] i am using -codec opy [04:46:06 CET] copy* [04:47:30 CET] i meant does it copy slightly lossy from original* [04:48:28 CET] pinPoint: it copies [04:49:23 CET] it puts the exact same streams in a new container [04:52:04 CET] furq: but does the stream loosy bitrate? [04:52:52 CET] no [04:52:54 CET] that wouldn't be a copy [04:52:58 CET] i went from what looks like 7900(Duration: 00:10:34.53, start: 0.000000, bitrate: 7980 kb/s) vs Copy (Duration: 00:10:34.60, start: 0.067000, bitrate: 7499 kb/s) [04:53:41 CET] i'll have another go [04:54:07 CET] if it's actually lost bitrate then you'll have noticed ffmpeg taking ten times as long to reencode it instead of just copying it [04:54:33 CET] it will also say in the output [04:55:10 CET] that is the result of ffprobe. But I will experiment again [04:55:12 CET] if you muxed it into a different container then i expect one of them is just reporting the bitrate wrongly [04:55:40 CET] i did not specify containers just -c:v [04:56:11 CET] ffmpeg -i bbb_sunflower_2160p_30fps_normal.mp4 -c copy -an BBB_COPY_NO_AUDIO.mkv [04:56:18 CET] yeah I changed container.... heehhe [05:43:58 CET] still getting that 7498 kb/s [05:44:08 CET] ffmpeg -i bbb_sunflower_2160p_30fps_normal.mp4 -c copy -flags global_header -an BBB_COPY_NO_AUDIO.mp4 [05:47:50 CET] "Duration: 00:10:34.53, start: 0.000000, bitrate: 7980 kb/s" is the description of the whole file, not just the video stream [05:49:06 CET] the video stream bitrate will be in the line starting with "Stream #0:0" [05:49:21 CET] depending on the container [06:44:56 CET] ah [06:45:03 CET] it was missing the audio i see. ok [13:36:41 CET] Hello, im trying to decode a video by writing bytes to the input stream of ffmpeg process in C#. [13:37:20 CET] By using the bytes i got and writing them to file, and afterwards making ffmpeg read from them, i get it to work. The problem is using bytes directly to supply ffmpeg. [13:37:31 CET] https://gist.github.com/Kwoth/2d5cf0301b533b647d01 this is what i try to do [14:09:28 CET] how can I debug avcodec_find_encoder(AV_CODEC_ID_H264); not finding the encoder? I have built with both --enable-libx264 and --enable-encoder=libx264 flags but I'm out of ideas ... [14:13:24 CET] have you registered all codecs? [14:13:56 CET] how do you mean? [14:15:06 CET] actually I think it linked against libx264.so instead of .a and when deployed on android it might have failed to load the .so, now I removed the libx264.so and ffmpeg configure does not find libx264 anymore :P [14:15:06 CET] there is function [14:15:37 CET] durandal_1707: I'm trying to follow the encoders example [14:15:46 CET] didn't see such function [14:16:07 CET] duh! [14:16:24 CET] I missed avcodec_register_all(); :P [14:16:53 CET] still if the .so of lib264 is available it seems to link to it, so it might fail I guess [14:32:54 CET] dantti, did you load the x264 so_ [14:32:55 CET] ? [14:33:31 CET] I was trying to static link everything [14:33:53 CET] now I get undefined refs.. to x264 calls [14:34:19 CET] Did you compile x264 for Android as well? [14:34:22 CET] yup [14:34:28 CET] Also why aren't you using MediaCodec? :) [14:34:34 CET] I was even calling it directly [14:34:39 CET] what's it? [14:37:33 CET] Interface to the device hardware H.264 encoder. [14:37:42 CET] Mavrik: can't that be used to generate a mpegts stream? [14:37:46 CET] x264 will be very very slow on those ARMs :/ [14:38:00 CET] Those are two different things. [14:38:16 CET] yup, it's the muxer part right? [14:38:19 CET] You can use MediaCodec to encode AAC / H.264 and then use something else to mux it into MPEG-TS. [14:38:24 CET] Mhm :) [14:38:38 CET] It's true that Android doesn't include a MPEG-TS muxer. [14:38:54 CET] But I'm just trying to warn you about the x264 performance - a 5 sec clip at 720p took up to 30s or more to encode on most Android devices. [14:39:24 CET] sure, I had to reduce image quality to not drop frames [14:39:46 CET] You will also drain the battery. [14:39:50 CET] I guess I need to learn how to call java code... [14:40:23 CET] On the other hand, the HW video encoder quality is significantly worse. [14:40:41 CET] You need to encode at 5Mbit what x264 would easily do in 1,5Mbit :) [14:41:48 CET] the problem is also writing java code is boring... [14:42:16 CET] BtbN: for the baterry the app is on a always on tablet so no issue there [14:42:31 CET] good morning everyone.. running CentOS Linux 7 (Core) ... using nvenc for encoding .. when i turn up an encoding session i start seeing errors like " http://pastebin.com/G0hcWZwG ".. also i see a bunch of pixelation on the screen. Anyone know why ? Cpu's dont' look like they are doing much but it is as if there some sort of limitation that is causing these errors. [14:43:35 CET] Those are decoder errors, your source is already broken. [14:44:35 CET] Mavrik: I can use more cores also to improve performance? of for muxing in mpegts it wouldn't work? [14:45:07 CET] x264 will use as much cores as it can get, muxing itself isn't really expensive CPU-wise [14:45:22 CET] the video encode takes a huge chunk of CPU [14:45:59 CET] ok I was asking because when I called x264 directly I was wondering how to distribute frames to encode and still get them at the right order [14:46:22 CET] as the call was blocking I didn't know how it worked [14:46:51 CET] hmm, x264 should return frames in DTS order always [14:47:10 CET] but ordering the frames is why DTS timestamp exists on them :) [14:47:35 CET] BtbN , i am using another encoder ( elemental ) and it seems to be outputing correctly from the same source. :( .. only ffmpeg does this [14:48:09 CET] i am trying to replace it with nvenc as it is more effecient but.... any way to correct some of these errors ? [14:48:19 CET] This is not related to the ffmpeg encoder at all, those happen wile ffmpeg is unpacking and decoding the h264 input. [14:48:24 CET] right I guess what I mean was how to give more frames as the encode function blocked, I wasn't sure If I'd need to create more thread by myself [14:55:18 CET] anyone know a good decoding filter that can help me get rid of those errors ? not sure which ones to use [15:03:30 CET] BtbN ? can you please help me point in the right direction.. i am stuck :( [15:03:36 CET] you usually pretty good .. [15:03:39 CET] Hi all :) [15:04:04 CET] Your source gives you a broken stream, fix your source. [15:04:27 CET] i can't fix it.. comes from satelite carrier [15:04:37 CET] and i am using an elemental encoder and i have no errors [15:04:47 CET] but with ffmpeg .. it acts up a lot [15:05:13 CET] Sounds like terrible signal quality. [15:05:21 CET] well its 720p but.. [15:05:27 CET] actually its 1080i [15:06:25 CET] Hope you can help me out, Does any one know what is this error means? ERROR: librtmp not found using pkg-config basically librtmp.so is saying is missing but its there. i have even build it from fresh again and still wont go any futher, openssl its built from source, rtmpdump too, but have problems with ffmpeg, can i count on your help anyone? [15:06:52 CET] or 720p .. depending.. we are taking the system from sat... outputing to UDP .. then taking that and transcoding it.. works fine on the other encoder and i am sure there are errors there too but at least the other encoder handles it well.. not sure how [15:07:03 CET] maybe there is some decode filter that is being used that cleans it up and then it doesn't mess up the output [15:07:13 CET] i dont know what filter to use :( [15:09:09 CET] There is no filter that magicaly fixes a broken stream [15:09:56 CET] ya i know.. but the output is there.. but i see pixelation like weather related issues or something .. i don't see it on the other encoder ( same exact source ) [15:10:33 CET] also. if i do 1 encode at a time it seems to be a lot better.. soon as i start pushing the server a bit it seems to act up more .. like cpu power but i am not using any cpu .. very very strange [15:16:42 CET] NoviceAlpha2: upload your config.log to a pastebin service and link here please [15:22:07 CET] BtbN .. i think this is something to do with some processing power or something.. so i encoded 1 time... no issues and no errors.. i add 4 simultanious encodings and i get errors flying on the screen.. like cpu can't keep up and causes errors.. [15:22:19 CET] but.. cpu's don't look like they are doing much.. i don't get it [15:28:36 CET] c_14: i am trying to change path's to toolchains, sdk, and libs, give me 5 min will see if script might be corrupted, im using oryginals from Compilation Guide Compile ffmpeg for Android with rtmp git clone. 1 min please [15:36:05 CET] c_14: there is pastebin link, http://pastebin.com/GQBUaHsN [15:38:16 CET] NoviceAlpha2: not build.log, ffmpeg's config.log. Should be in the ffmpeg source directory [15:39:36 CET] c_14, will do 1 sec. [15:40:05 CET] You probably just have to set PKG_CONFIG_PATH to "/home/test/android-ffmpeg-with-rtmp/src/rtmpdump/librtmp/android/arm"/lib/pkgconfig though [15:41:08 CET] c_14: http://pastebin.com/H9jW1q4W [15:41:26 CET] Package 'libssl', required by 'librtmp', not found [15:43:52 CET] c_14: thanks i did not knew that config.log is there, ok will try to apply missed liblaries, thanks a lot [15:49:37 CET] hey i may be onto something here.... source is h264 already.... i am doing " ${ffmpeg} -y -i "$stream" -c copy -map 0:p:$6 -f mpegts - | tee .... " what should i use instead of mpegts so i dont convert .. i just want to pass on to the tee to have the other stuff encode it [15:50:43 CET] Does someone here have experience with sony vegas? Or do you guys know a channel for sony vegas questions? [15:50:54 CET] rawvideo !!! [15:52:29 CET] mpegts is a container. if you used -c copy then it doesn't matter what you specify with -f (as long as it's acceptable for h264) [15:59:42 CET] oh.. then my problem is not solved [15:59:43 CET] :( [16:00:00 CET] basically just want to output the input and tee it off to multiple processes :( [16:07:39 CET] Do you guys think there would be any visible difference between tga -> libx264 crf 4 -> editing/lossless .avi -> libx264 crf 18 and changing crf 4 to crf 16 in the first step? [16:18:49 CET] it'd be pretty hard for a human eye to notice without frame-stepping the video. but it could be [16:19:03 CET] but during high motion scenes it's far more likely [16:19:48 CET] if you view the high-motion scenes in slow motion or frame-by-frame, at least :) [16:20:38 CET] if you want constant quality, you use -qp instead. Using -crf will still reduce the image quality for high motion [16:26:46 CET] of course, if you have the disk space and bandwidth, you could just skip the first x264 encode and put it right into the lossless format for editing... [16:29:24 CET] well if there is an intermediate lossless step then it should be... [16:35:20 CET] kepstin_: The problem is that the vegas preview starts lagging horribly with crf 4 already, with crf 16 it's fine. [16:35:51 CET] DHE: I thought crf is already a variable bit rate/constant quality? It doesn't work the same as -qp? [16:36:42 CET] crf is approximately constant visual quality (uses psy optimizations) when played back realtime [16:39:45 CET] -qp is static quantizer, which means constant mathematical quality (this isn't exactly true, since there's still boosting on keyframes, etc) - but it's not really constant visual quality to a human viewer. [16:40:07 CET] close tho [16:42:31 CET] Hmmm, I guess crf still makes more sense then^^ [16:43:19 CET] keep in mind that the psy optimizations in crf mode are explicitly designed to remove data that might not be noticed in the video as it currently is. [16:43:30 CET] if you're doing editing afterwards, that probably doesn't make sense [16:44:32 CET] Hmm yeah maybe. Does -preset still work the same way with -qp? [16:50:26 CET] Maybe -tune fastdecode would help, too [17:03:11 CET] qp 12 and fastdecode seem to work fine, preview is like 20 fps but that's enough for editing^^ [17:10:44 CET] hi, does ffmpeg support googles tinyalsa as an replacement for alsa? [17:11:45 CET] refering to the alsa capture device support in libavdevice [17:17:15 CET] Afaik no [17:26:10 CET] the encoder/decoder files for alsa doesn't seem too intimidating, might be I can implement a tinyalsa device. has this been attempted? [17:29:05 CET] hey guys.. question.. i have a transport stream with many program ID's [17:29:40 CET] on of with is program 1895 for example.. i use " -map 0:p:1895 " to pull that out .. problem is the stram 0:19 ( video ) and the audio streams keep changing [17:30:23 CET] how do i map all the audio i want while the stream keeps changing [17:30:30 CET] can i somehow select the " 0x1511 " [17:31:41 CET] hanDerPeder: As I understand both alsa-lib and tinyalsa are just libraries providing an API to talk to alsa kernel drivers. So wherever tinyalsa works - alsa-lib should work too. [18:06:52 CET] hey anyone know how to select streamid hex inside a transport stream program id ? [18:07:17 CET] eg.. --map 0:p:$programid .. then -map 0xxxxx 0x2222 etc etc.. for the actual stuff i want from that program ID [18:11:52 CET] prelude2004c: https://ffmpeg.org/ffmpeg.html#Stream-selection [18:24:15 CET] hello, first of all thanks for this incredible application, its fantastic, i've been experimenting a bit with it and its great [18:24:57 CET] hey, can someone help with with this >>> http://pastebin.com/qSCgYKDm [18:25:03 CET] i am doing something wrong [18:25:07 CET] by any chance is there a site or blog or something showing some cool tricks with it? [18:29:34 CET] the trac web site has a couple of things strewn across various pages, from encoding tips/tricks to hard-to-use features to visualizing the video encoding itself [18:35:12 CET] DHE -> thanks, one more question, in the command line how is the presedence of the commands considered? i am taking like 10s from a video-resize it and convert it to animated gif, that is done perfectly but now i tried to flip the gif horizontally, so i added -vf "hflip" just before the output name of the gif, but now it did something totally different, it didnt resize de video, showed just a part of i [18:35:18 CET] t and made a lot much bigger file, ... [18:35:21 CET] ... from 2.3M to 16M [18:37:40 CET] the format of options are: ffmpeg [input1 options] -i input1 [[input2 options] -i input2 ...] [output1 options] output1 [[output2options] output2 ...] [18:38:01 CET] prelude2004c: 0:p:1895:0 and 0:p:1895:1 afaik [18:38:29 CET] now I'm not 100% sure but using -s will result in a scale filter being used to resize the video. maybe mixing -vf (manual video filters) affected that [18:38:37 CET] http://pastie.org/10592871 [18:38:55 CET] ah... yeah, chain your filters together. [18:39:04 CET] -vf scale=600:-1,hflip [18:39:45 CET] i see, i just c&p and didnt see that im repeiting -vf , so dumb, thanks man [18:42:07 CET] libavcodec.a has x264_picture_init Undefined, if the link is static shouldn't it be defined? [18:44:09 CET] libx264 needs to be linked as well [18:46:08 CET] DHE: on my app? [18:47:06 CET] if so I did that, actually I can get rid of undef symbol if I call the function in my app, but then there is another 10 calls to x264 lib [18:49:19 CET] I'm saying x264_* functions require linking to libx264 [18:55:10 CET] it's a static build I'd think that if libavcodec.a compiles fine it got linked but when I try to link to it I get the undef for libx264 [18:55:39 CET] dantti: yeah, because you have to also link to libx264 to pull in the code for it. [18:55:39 CET] dantti: -Wl,--whole-archive [18:56:26 CET] static libraries don't link anything, they're just an archive of object files [18:57:15 CET] furq: that's probably why ldd doesn't return a link list as does .so file :P I add that to --extra-ldflags right? [18:57:24 CET] and static libraries don't know anything about dependencies, so you have to specify all of them during the final link. it's not like dynamic libs where linking to one pulls in its dependencies automatically. [18:57:50 CET] -Wl,--whole-archive -lx264 -Wl,--no-whole-archive [18:58:16 CET] when compiling your app [18:58:45 CET] ok, I thought it would be when compiling ffmpeg [18:59:07 CET] I have -lx264 in my app just not this (no)-whole-archive [18:59:15 CET] btw don't they conflict? [18:59:17 CET] I don't think you need the whole-archive stuff, just putting -lx264 after the libav* libraries should be sufficient for the linker to pull in the needed stuff. [18:59:21 CET] no [18:59:35 CET] whole-archive applies to all static libraries until the next no-whole-archive [18:59:41 CET] kepstin might be right though [18:59:47 CET] with static archives, the order of linking is important :/ [18:59:50 CET] must be after? I noticed that ffmpeg *.a need a right order [19:00:31 CET] it's been a while since i linked anything with static libraries because stuff like this happens [19:02:10 CET] hey ,can anyone point me in the right direction >> http://pastebin.com/qSCgYKDm [19:03:27 CET] what gets really fun is when you have two static libraries that use functions from each-other. e.g. if liba uses a function from libb and vice-verse. Then the correct link command is "-la -lb -la" :/ [19:03:36 CET] (or use -Wl,--whole-archive) [19:04:35 CET] surely the first item in the correct link order is "go back and make them shared libraries you fool" [19:05:03 CET] kepstin: oh, I was trying to do something like that a while ago [19:05:12 CET] decided to merge the two :P [19:06:23 CET] basically, the way it works is that the linker has a list of symbols in the program that it's missing the code for. Each time you pass a -l option with a static archive, it looks in that archive, and pulls in the functions that satisfy missing symbols. Then, after that, it adds any new missing symbols from the new functions to its list, and goes on to the [19:06:23 CET] next library. [19:07:53 CET] what the "whole archive" option does is makes it copy *all* of the functions from the static archive into the program, instead of only the functions that satisfy missing symbols. [19:07:54 CET] so the -la -lb -la is just for a "second scan" of missing symbols right? [19:08:07 CET] yeah [19:09:04 CET] yay, it compiled now let's see find_codec works on android this time :P [19:11:04 CET] great, I can now start coding :P [19:11:07 CET] thanks [19:28:40 CET] anyone have a sec to help me with a map function ? [19:30:04 CET] Stream #0:12[0x1511] the stream 0:12 keeps changing everytime i run it.. but the 0x1511 always maintains [19:30:21 CET] how do i use a selection based on the " 0x1511 " [21:06:02 CET] im reading the -ss input/output and i dont understand really what the difference is, or when to use one or the other, for what i understand its better and faster to use input right? before the -i ? [21:16:17 CET] autofsckk: before input the seeking is done via demuxer, before ouptut the seeking is done via decoder (I think). anyways, The latter is more accurate, but you lose speed. [21:17:41 CET] ok ,thanks for the explanation, i dont know what demuxer and decoder are, well i have an idea but im really new at video editing, ill look about it thanks again [21:18:26 CET] autofsckk: demuxer helps you get the video/audio stream contained in a file. Decoder does decoding of the coded video/audio stream [21:19:42 CET] ok, thanks [21:38:06 CET] anyone knows a tools that prints dolby leq(m) loudness ? [21:48:57 CET] autofsckk: basically, when -ss is an input option, ffmpeg tries to jump to the correct spot in the file and starts decoding there. When -ss is an output option, ffmpeg decodes the entire file from the start, but just throws out frames until it gets to the spot you wanted. [22:33:29 CET] Hi! I'm building a music analyzer, but I totally forgot about the floating-points formats in my decoding process. Is is really dirty to get floats in a temporary variable, like this v = ((float*)decoded_frame->extended_data[0])[i];, and then multiply it by max_in16_t, so that I can work with a � normal ��s16 array? [22:35:49 CET] lossy audio in general is natively float, and only those formats by default get float output [22:35:54 CET] heh, i think it's more common for people to convert the int formats to floating format as a working format rather than the other way around. [22:36:16 CET] there is avresample and swresample to convert float outputs to integer if you require that [22:36:27 CET] take great care one is 32 bit and the other is a signed 16 bit [22:36:51 CET] Hi, are -probesize and -analyzeduration the only ways to reduce startup latency when re-streaming via ffmpeg? I'm using filter_complex to turn 6 security cameras from a black box DVR that streams via rtsp and piping that to mpv to view locally [22:36:52 CET] use swresample to produce your target format [22:36:56 CET] the decoders themselves output whatever is native to them, basically [22:37:15 CET] Is there no way to specify the parameters that the source streams use beforehand? [22:37:22 CET] what [22:37:26 CET] oh [22:37:29 CET] go on [22:37:31 CET] Primer: you can try the buffer "reduction" settings, like nobuffer [22:37:39 CET] okay, so you suggest that I look into swresample? [22:37:41 CET] hey, i am trying to remove an audio channel from a dual audio video file, the original sounds fine but when i use the following command the audio goes severely out of sync -- ffmpeg -i in.avi -map 0:0 -map 0:2 -acodec copy -vcodec copy out.avi [22:37:44 CET] I'm not sure how well that works [22:37:49 CET] wouldn't it be a loss in term of performance? [22:38:06 CET] ChocolateArmpits: Well, I'm getting this: [cache] Cache is not responding - slow/stuck network connection? [22:38:17 CET] well optimized resampling is going to be quicker than unoptimized (if you require resampling) [22:38:19 CET] Which lead me to looking at options to disable any form of caching [22:38:40 CET] there are two APIs with similar APIs (av|sw)resample available in FFmpeg, feel free to pick your poison [22:38:44 CET] Polochon_street: swresample is a pretty well optimized chunk of code; you'd use it to convert a whole buffer at a time between formats. [22:38:56 CET] okay, I trust you, thanks very much! :) [22:39:35 CET] basically you are going to get integer based output for lossless formats generally, while for lossy you are going to be getting floats [22:40:24 CET] or well, for lossless you are going to get whatever there was in the input [22:42:06 CET] yep, I only use lossless formats, and when someone tested what I wrote with a .mp3 well... It broke everything [22:42:16 CET] I'll be more cautious next time! [22:42:36 CET] can anyone help me with my audio sync issue? [22:43:46 CET] I see no such option as nobuffer...unless you mean mvp [22:44:20 CET] I've been messing with -probesize and that's made the video start much more quickly than without it [22:44:39 CET] but ffmpeg is still reading each source stream in sequence [22:45:19 CET] well, it _seems_ it's reading them sequentially [22:45:47 CET] Primer: -fflags nobuffer [22:45:51 CET] it's an input flag [22:46:17 CET] https://www.ffmpeg.org/ffmpeg-formats.html#Format-Options [22:47:16 CET] Anyhow, this is what I'm doing: http://ix.io/mzg [22:48:32 CET] It's taking about 5-6 seconds before it appears [22:48:50 CET] and I'm trying to get that down to as little as possible [22:49:04 CET] then reduce analyzeduration [22:50:11 CET] try something like 500 000 microseconds, which would be half second [22:50:29 CET] lower it until ffmpeg process input anymore [22:50:33 CET] can't* [22:51:21 CET] The output from ffmpeg suggests it's reading each source sequentially. I don't suppose there's a way to make that concurrent? [22:52:16 CET] Also, if the source is consistent, can't the source "attributes" (the things probing and analyzing are looking for) be specified beforehand? [22:52:55 CET] I would guess the input are analyzed parallelly, but I'm not sure [22:53:10 CET] you can specify framerate, format [22:53:25 CET] again, I can't say if this will speed up the analysis [22:53:39 CET] you can specifiy resolution and pixel format too [22:54:08 CET] anyways start with analyzeduration [22:54:29 CET] if there really are latency problems this is a definit setting that can change that [22:54:35 CET] I set that to 0. The stream continues to work, but there's no appreciable speed up [22:55:08 CET] ok try with one input [22:55:09 CET] Well, I'm remote from the source at the moment, but I'm using the same script when I'm not [22:55:15 CET] and don't set something to 0 [22:55:24 CET] which is then on a local gigabit network [22:56:01 CET] default for -analyzeduration is 0... I wonder if that has the effect of 'unlimited'. [22:56:08 CET] right [22:56:22 CET] set it to a sensible value [22:56:28 CET] as I said 500000 [22:57:15 CET] yup, tried that, no difference [22:57:27 CET] Primer: for all the inputs ,right ? [22:57:33 CET] piping ffmpeg | mvp is hosing the terminal [22:57:34 CET] yes [22:57:47 CET] http://ix.io/mzh [22:58:13 CET] Is the latest version [22:58:31 CET] you should not be using probesize when using analyzeduration. The idea behind them is the same, but you either alter seconds or bytes [22:58:48 CET] ahh [22:59:01 CET] basically pick one depending on your situation [23:00:22 CET] also mpv may have it's own buffer, see if you can reduce it there. Or just use ffplay and transplant latency reduction settings there [23:01:40 CET] ffplay chokes on this [23:01:42 CET] err [23:01:43 CET] no [23:01:56 CET] its fullscreen option is real full screen [23:02:01 CET] not a borderless window [23:02:11 CET] which is totally unacceptable [23:02:12 CET] is that even an issue? [23:02:16 CET] yes [23:02:22 CET] I have 4 displays [23:02:24 CET] afterall you're testing for latency and not for preview [23:02:48 CET] and the displays flicker on and off for several seconds while going real full screen [23:03:15 CET] is the latency present when you only have one of the inputs passed ? [23:03:51 CET] yeah, -analyzeduration 500000 just goes back to taking forever for each stream [23:03:58 CET] which is being handles sequentially, for sure [23:04:23 CET] Not sure what you mean [23:04:41 CET] when you only have one single input, rather than all of the them [23:05:03 CET] can you "ffplay rtsp..." ? [23:05:12 CET] and check what's the latency with just one of them [23:05:23 CET] without all the added filter processing [23:06:15 CET] seems ffplay is a separate package [23:06:43 CET] but if I use mpv with just the rtsp url it takes several seconds [23:07:19 CET] I have ffmpeg installed as a package (Linux Mint 17) but I compiled it myself. Not sure if ffplay was a separate package [23:07:54 CET] mpv has options to set libavformat parameters [23:08:00 CET] see the documentation or poke #mpv [23:08:19 CET] at least I remember it having a way to pass lavf parameters :P [23:08:24 CET] vlc URL == 1-2 seconds [23:08:36 CET] from hitting enter to where I see output [23:08:54 CET] because it doesn't use libavformat? [23:09:00 CET] at least for most formats, by default [23:09:18 CET] you can make it use by switching the demuxer on command line [23:09:49 CET] not sure if mplayer is even relevant any more, but it just chokes [23:10:03 CET] it's barely being maintained by a guy or so [23:10:37 CET] probesize seems to be the most effective argument so far [23:11:17 CET] https://mpv.io/manual/master/#options-demuxer-lavf-o [23:11:19 CET] So going back to the entire reason for probesize and analyzeduration...can I not just specify these things? [23:11:21 CET] or heck [23:11:33 CET] mpv has demuxer-lavf-probesize [23:11:40 CET] Because the source format/framerate/etc. is not going to change [23:11:43 CET] and demuxer-lavf-analyzeduration [23:11:57 CET] some of the things can be defined, yes. some can't [23:12:03 CET] at least not on command line :P [23:12:27 CET] Well, the sources are all from the same device, so they'll all be identical, which the exception of stream 1 having audio [23:12:34 CET] and the others having none [23:13:12 CET] As far as duration, they're never ending streams, so... [23:13:33 CET] All that remains are...transport, codecs, bitrates? [23:13:58 CET] well you can set the format at least, and the decoder. it will still have some parsing requirements [23:14:04 CET] format as in the container [23:14:47 CET] I also have this: [nut @ 0x29a0ca0] Non-monotonous DTS in output stream 0:1; previous: 51611, current: 51608; changing to 51612. This may result in incorrect timestamps in the output file. [23:15:06 CET] This seems to cause camera 1, the one with the audio, to lag behind the other 5 [23:15:23 CET] video lag, that is [23:19:03 CET] Anyhow, I'm very satisfied with this as it stands right now. Just trying to refine it as best as I can. [23:19:37 CET] well, did you try poking at the mpv options I noted? [23:19:38 CET] At home I'm running a 50' HDMI cable from the device to the TV in my living room. The direct output from the device is real time [23:20:12 CET] I have not, but from what I can tell it seems the majority of the the delay is happening on the ffmpeg side [23:20:13 CET] basically analyzeduration and probesize and you can define the container [23:20:26 CET] well those settings control libavformat... [23:20:32 CET] which is why I noted them from the goddamn manual [23:20:42 CET] Pardon my ignorance, but how would I determine the container? The presume it's in the output? [23:21:12 CET] container is called "format" in ffmpegspeak [23:21:24 CET] I see nut mentioned in that one line you quoted, for example [23:21:39 CET] in ffmpeg it's around the "Input 0/1 things :P" [23:22:44 CET] well, nut is what I'm using for output [23:23:03 CET] well then you define that in the reading side [23:25:11 CET] sure [23:25:41 CET] The sequence of events have lead me to try to speed up the ffmpeg side of things first though [23:26:14 CET] initially without probesize, I'd see the output from the first stream take several seconds [23:26:35 CET] again for the next, next, etc, then after a good 20-25 seconds, the video would appear [23:26:59 CET] now with -probesize on the ffmpeg side, each video's output happens in about 1 second [23:27:16 CET] http://ix.io/mzl [23:27:21 CET] that's the output [23:27:33 CET] redacted to remove the private parts [23:28:49 CET] So I type ./ffmpeg.sh, hit enter, a chunk of text appears from reading the first stream in the first second, then the next check the second after that, etc. etc...6 seconds after hitting enter, the video appears [23:29:11 CET] yeah, the ffmpeg tool sequentially probes each input in order before setting up the filters and outputs. No real way around that. [23:29:27 CET] I was hoping there was a way to do that concurrently [23:30:03 CET] but I'd settle for speeding up the "probing" by specifying all parameters beforehand, considering the stream parameters (attributes, whatever) won't ever change [23:30:49 CET] And I suppose I'd also try speeding up the mpv end as well by specifying [23:31:11 CET] thing is, that's entirely local, so I figure that's already pretty optimal [23:31:26 CET] the mpv end, that is, since I'm piping to it [23:31:31 CET] I wonder if predownloading the sdp to files and passing those to ffmpeg might help [23:31:41 CET] since it's the sdp that has all the connection setup stuff [23:32:14 CET] hmm, but I guess rtsp does more than that :/ [23:33:01 CET] tbh, if you want 0 setup time, directly using rtp with the server sending to statically configured addresses/ports would do it. [23:35:11 CET] I suspect that most of the per-stream setup time is the rtsp negotiation, rather than any stream probing. [23:45:18 CET] if you have to stick with rtsp, it might be that the best option is to set up the rtsp connections in an external app (in parallel), write the sdp to disk, then start up ffmpeg with the sdp. [23:45:28 CET] yeah, it's a black box [23:46:05 CET] I know it's running Linux and the telnet port is open, but I haven't looked into hacking it [23:46:14 CET] I'm guessing it's using hardware encoders too [23:46:28 CET] since the PC board on it is minimal [23:46:47 CET] the sdp would contain all the codec and format information, so in that case ffmpeg is basically just opening some udp ports for the rtp/rtcp and getting on with it. [00:00:00 CET] --- Tue Dec 1 2015 From burek021 at gmail.com Wed Dec 2 02:05:01 2015 From: burek021 at gmail.com (burek) Date: Wed, 2 Dec 2015 02:05:01 +0100 (CET) Subject: [Ffmpeg-devel-irc] ffmpeg.log.20151201 Message-ID: <20151202010501.C7FD218A00AE@apolo.teamnet.rs> [00:07:08 CET] i cant find how to cut the "canvas", is it possible? i mean if i have a video 600x400 for instance, but i want to get rid of the border, like the ones made from iphones recordings and i want to make it maybe 350x400 , is it possible? [00:08:05 CET] kepstin -> thanks for the explanation [00:08:06 CET] Use a crop video filter. [00:09:27 CET] i see, this one http://ffmpeg.org/ffmpeg-filters.html#crop right? [00:10:22 CET] reading it, thanks [00:28:22 CET] why the hell doesn't -to actually work when cutting?? [00:28:29 CET] am I missing something super obvious [00:28:42 CET] i hate having to go to wolframalpha to do dumb shit like subtract timestamps [00:29:22 CET] -ss 1:52:22 -to 2:06:34 doesn't work, but -ss 1:52:22 -t 14:12 does [05:32:58 CET] ... wow [05:33:02 CET] thanks for all the replies brors [08:18:32 CET] Suppose I'm receiving some audio/video streams and encoding to mp4(h264,aac) on the fly; is there a way to put the moov atom at the start of the file so players can start playing it when it's only partially processed? [08:19:07 CET] I've seen -movflags faststart but comments on the site I found it suggest that it only works as a second pass [08:20:06 CET] Is it fundamentally impossible to write a (complete) moov atom without having seen the entire file? [08:38:40 CET] wlritchi: hallo [09:03:16 CET] Hi everyone, I'm trying to follow this post http://blog.oscarliang.net/webcam-streaming-video-raspberry-pi-via-browser/... but although I can open the stream from VLC when i try to open it from web browser (chrome) it dowloads mjpeg, i've read around that changing the port might help, to be honest havent tried that, anymore suggestions? [09:30:26 CET] maco: hallo [09:30:43 CET] wlritchi, it's fundamentally impossible to write the MOOV atom before you write data (it contains an index) [09:31:00 CET] wlritchi, I suggest you use MPEG2-TS if you can or use MP4 segmented mode if your player supports it [09:36:15 CET] hi waressearcher2 [09:50:02 CET] maco: wie geht's es dir ? [10:04:28 CET] Mavrik: Thanks for the advice; I'll look into other transports since that's probably easier [10:17:18 CET] @waressearcher2: Im fine thanks, No german here BTW! xD [10:26:41 CET] hallo, anyone here experiengin issues when streaming mjpeg it wont open on browser (chorme) instead it download a file.... [10:45:38 CET] howdy [10:45:47 CET] xintox: hallo [10:52:35 CET] ye'llow [11:08:17 CET] man. debian version on wheezy is totally outdated [11:08:31 CET] does ubuntu have an easier way to get latest versions of ffmpeg? [11:10:41 CET] xintox, get static builds. [11:11:11 CET] xintox: http://johnvansickle.com/ffmpeg/ [11:15:39 CET] oy [11:15:58 CET] xintox: #ubuntu ? [11:16:28 CET] wtf is .tar.xz file? [11:28:13 CET] nm [11:28:20 CET] i am getting old [11:33:23 CET] :-) [12:01:48 CET] hi is there anyone here?!?!?!??!? [12:01:50 CET] :S [12:01:57 CET] does hls_time have to coincide with source stream? [12:02:08 CET] maco: hey [12:15:01 CET] Hi, [12:15:36 CET] I created a mjpeg stream it open fine on VLC but when opened from browser it download the mjpeg file... [12:15:38 CET] ? [12:18:59 CET] maco what browser? [12:23:31 CET] chrome [12:23:34 CET] and IE [12:50:31 CET] Action: maco slaps xintox around a bit with a large fishbot [12:50:41 CET] xD [13:24:06 CET] xintox: you can set hls_time to whatever you want, but keep in mind that the keyframe interval will regulate it [13:24:57 CET] I find 4s to be a good minimum, and 10s as a common default or maximum by the industry [14:04:17 CET] I'm creating an AVFrame from a QVideoFrame, since QVF has the image bits mapped I'm just assigning to the avframe->data[] planes, but I'm afraid of when avframe get freed it frees that data, does this look sane? https://paste.kde.org/pvvhg0bq7 [14:06:21 CET] Are you sure the pix_fmt is compatible? [14:07:02 CET] Uhh, that probably won't fly well. [14:07:22 CET] Check the doc on what encode_frame guarantees around the buffer ownership [14:08:17 CET] BtbN: for now I hardcoded AV_PIX_FMT_NV21 which is what QVideoFrame->pixelFormat() returns (QVideoFrame::Format_NV21) [14:11:29 CET] Mavrik: you mean avcodec_encode_video2? if so it doesn't say anything about the frame [14:11:59 CET] since the code compiles if I write the output to a file will it be played somewhere (as I'm not muxing yet)? [14:13:10 CET] *playable [14:25:27 CET] the code runs at least hopefully it ain't leaking... :P [14:57:16 CET] 3rd attempt... [14:57:18 CET] Hi everyone, I'm trying to follow this post http://blog.oscarliang.net/webcam-streaming-video-raspberry-pi-via-browser/... but although I can open the stream from VLC when i try to open it from web browser (chrome) it dowloads mjpeg, i've read around that changing the port might help, to be honest havent tried that, anymore suggestions? [14:58:08 CET] THE COMMAND... [14:58:08 CET] ffserver -f /etc/ffserver.conf & ffmpeg -v verbose -r 5 -s 640x480 -f video4linux2 -i /dev/video0 http://localhost/webcam.ffm [15:03:47 CET] i don't think browsers support mjpeg [15:11:07 CET] any format that browser would support? strange becuase the link I gived at the top said that it worked for him??? [15:28:08 CET] ok [15:28:13 CET] found tha issue at last [15:28:14 CET] !!! [15:28:20 CET] FYI* [15:28:37 CET] bug with the latest version of Chrome [15:28:45 CET] [15:28:47 CET] ... [16:20:53 CET] Do I need this line? DATA_FILES := $(wildcard $(SRC_PATH)/presets/*.ffpreset) $(SRC_PATH)/doc/ffprobe.xsd [16:21:15 CET] I'm only using this library to get net streams and decode into pixels [16:21:26 CET] looks like some sort of shaders? [16:53:42 CET] hi. i am trying to encode a h264 file from a device to file and play it in a html5 video element at the same time. the recording goes fine, the playing works too, but the browser wants to load the entire file before it can play. i already use -movflags frag_keyframe+empty_moov but for some reason this doesn't seem to work. anyone has an idea? [16:58:30 CET] Hello [16:59:37 CET] I'm using ffmpeg libraries and was wondering: If I have a generic audio stream, not mono. How can I transform it in mono? [17:00:38 CET] swresample and setting the target to mono [17:01:49 CET] Franciman: http://stackoverflow.com/questions/14989397/how-to-convert-sample-rate-from-av-sample-fmt-fltp-to-av-sample-fmt-s16 [18:02:55 CET] If I'm using ffplay to just decode h264 for pixel reads only Do I really need to configure filters? [18:04:47 CET] usually no [18:07:33 CET] So if i just open a stream rtsp i would need protocol rtsp [18:07:40 CET] and probably h264 if its encoded like that [18:08:45 CET] As far as muxers and demuxers go... im reading some of the docs and they seem to be audio related.... but the ffmpeg.org says demuxers are configured elements that can read the multimedia streams which seems to indicate they deal with video as well [18:09:03 CET] I dont need anything audio period... what -disables would i want to use to ensure this? [18:16:12 CET] Good evening. I just upgraded to 2.8.3, and now my tried and true command line doesnt run anymore. http://pastebin.com/wt9XYbgN [18:16:32 CET] Cannot select sample rate for the link between filters audio format for output stream 0:1 and output stream 0:1. [18:23:19 CET] hmbg: looks like a bug to me, open a bugreport on trac [18:23:40 CET] oh snap [18:23:54 CET] mention that it's a regression [18:24:04 CET] any suggestions for a workaround? [18:24:04 CET] although, could you test with a build from git first? [18:24:20 CET] I could try [18:27:06 CET] AS for a workaround, I'd try and minimalize the command so you know exactly where it's breaking (only use one output at a time, find which of the outputs is causing the error, find which filterchain is causing the error by removing them one at a time etc) [18:27:43 CET] Then probably add an aformat filter to that filterchain [18:31:06 CET] ok, will look into it [18:31:34 CET] just to double check, test with a build from git you mean checkout the latest from git and build it from source? [18:34:35 CET] either that, or use the static git build [18:48:54 CET] does not work with a git build [18:49:10 CET] Ok, then it's a bug [18:50:01 CET] Ill post it in trac [18:50:28 CET] Mention that it's a regression between whatever version you had before and 2.8.3 [18:50:36 CET] and paste the output from the git build [18:50:43 CET] (to the ticket) [18:52:51 CET] how about this... configure --disable-all-audio [18:52:57 CET] what does that equate to? [18:53:40 CET] There's no such option [18:53:56 CET] Ah, you want to disable them [18:53:58 CET] right so how would i accomplish this or is it not possible now [18:54:06 CET] yeah i only obtain a stream rtsp decode it [18:54:13 CET] do some shit with pixels and im done [18:54:34 CET] It's probably easier to disable-everything and enable the things you need [18:54:43 CET] Yeah im getting to that point [18:54:59 CET] You could of course list all muxers/demuxers/parses/decoders and encoders and disable all the audio ones [18:57:11 CET] btw the aformat line did the trick, so it works now. Thanks a bunch c_14 [18:58:36 CET] how well would running 4 ffplays connected to different streams work out on a intel i3? not so well? [18:59:18 CET] i ran about 4 on a i7 and it seemed to handle it but on another i7 it wasnt so good (slightly less specs) so been going through it trying every thing possible to make is thinner [19:18:53 CET] does ffmpeg automatically sets x264_opt_.i_threads ? [19:32:40 CET] dantti, yes [19:32:58 CET] X264_THREADS_AUTO is the default IIRC [19:33:38 CET] Mavrik: k, thanks, I'm now able to write a h264 file, now I need the muxing part :P [19:51:02 CET] It seems like --enable-lto causes assembly restraints linker errors [19:51:31 CET] using this line: [19:51:31 CET] ./configure --enable-shared --disable-static --prefix=/usr --arch=x86 --disable-postproc --disable-swscale-alpha --disable-ffmpeg --disable-ffprobe --disable-ffserver --disable-doc --disable-htmlpages --disable-manpages --disable-podpages --disable-txtpages --disable-encoders --disable-outdevs --disable-indevs --disable-muxers --disable-filters --disable-debug --enable-pic --enable-lto [19:59:53 CET] if i add --disable-asm to that it compiles... but dont i lose a ton of optimizations or do I really lose anything at all... [20:01:09 CET] awfulmakefile: you don't want to use --disable-asm, that'll make your ffmpeg a lot slower. [20:01:38 CET] and unfortuantely -flto is incompatible with inline asm right now. and inline asm is the better performance option compared to -flto [20:12:48 CET] so yeah anybody know why I have to --enable-lto --disable-asm on x86 processor? [20:12:58 CET] and I cant just go --enable-lto and call it a day [20:22:05 CET] shincode: you're the person previously known as "awfulmakefile"? [20:22:27 CET] shincode: you missed this response: and unfortuantely -flto is incompatible with inline asm right now. and inline asm is the better performance option compared to -flto [20:22:44 CET] Yeah [20:22:49 CET] I switched from one machine to next [20:22:56 CET] I got sick of waiting for lto to finish lol [20:23:04 CET] on a 2.1ghz dual core [20:23:24 CET] right now, on x86, lto provides negative benefit. so don't bother. [20:23:34 CET] for ffmpeg specifically, at least. [20:23:41 CET] So I see... I should knock off LTO and the asm optimization is implied [20:23:50 CET] as long as i knock off disable-asm as well lol [20:28:45 CET] it might be interesting to see if lto makes any difference when all the inline assembly is moved to other files, but making that change to ffmpeg would be a lot of effort and I dunno if it's even all possible. [20:29:17 CET] and in some cases it could mean adding additional function calls to assembly instructions that used to be inline... [20:31:53 CET] is the assembly supposed to be better than what compiler generates for optimizations? [20:32:06 CET] because it never seems to be the case on vstudio windows apps [20:36:14 CET] most of the assembly stuff that matters is for SIMD instructions (SSE/AVX), which compilers have traditionally been pretty poor at automatically generating. [20:36:33 CET] oh so it uses those... then yes. [20:36:37 CET] In general, new assembly code is only added if it can be demonstrated to be faster than the compiler generated code. [20:37:11 CET] i have to piece in h264 and mjpeg rtsp http for ffplay to use [20:37:17 CET] then disable strip the rest of it. [20:37:45 CET] so in order for rtsp to work i assume i have to enable... demuxer rtsp or something? [20:38:58 CET] to figure out what kind of thing rtsp is, the easiest way is often to just run "./configure" and see which header it's listed under [20:40:03 CET] (in this case, it appears to be a demuxer, yes) [20:40:35 CET] now here is a interesting one [20:40:43 CET] Im guessing i need to look for more h264's [20:40:54 CET] I just enabled h264 decoder and its whining :) [20:42:23 CET] in most cases, enabling a particular option will pull in its dependencies, but I guess that doesn't always work with stuff like rtsp where it can work with different codecs that have different requirements [20:44:57 CET] so is h261 h263 some older video formats before h264? [20:47:18 CET] yeah. h261 is more or less mpeg-1, and h263 is kind of a precursor to mpeg4 codecs like divx/xvid/msmpeg4. [20:47:39 CET] oh got it. looks like parsers for h264 or rtsp one of the two did not toggle on when enabling decoder h264, perhaps someone should report it :O [20:47:58 CET] nah, you can decode h264 without the parser depending on the container. [20:48:08 CET] Ah. [20:48:23 CET] the parser probably has to be enabled to get the connection between rtp and h264 working, i guess. [21:08:37 CET] hello, me again. I managed to decode audio from the web to pcm in real time but the quality is really poor and its choppy. I have decent pc (i5, 8gig ram, r9270x) [21:08:47 CET] these are the commands i use for running ffmpeg Arguments = "-i \"" + Uri.EscapeUriString(file) + "\" -f s16le -ar 48000 -af volume=1 -ac 1 pipe:1", [21:09:43 CET] basicaly if i could somehow buffer part of the song or something like that, that would be perfect, at least make it not be choppy. [21:10:26 CET] i tried with "-maxrate 960 -bufsize 1920" and "-maxrate 970 -bufsize 9600", nothing changes [21:15:05 CET] volume filter with 1 as art does nothing [21:27:29 CET] ok ill remove it but that doesnt fix my problem :d [21:34:26 CET] why you use ac 1? [22:06:23 CET] no idea what it is tbh. [22:06:49 CET] i took it from another lib that does audio decoding from stream [22:06:53 CET] that sets output channel number [22:07:11 CET] ah yes [22:07:14 CET] i remember now. [22:07:16 CET] I need mono. [22:07:47 CET] The c# library i am using does not support sending stereo sound to the voice channel. [22:14:57 CET] basicaly what i am doing is getting stream from youtube real time, decoding it to PCM, and sending those to the voice channel in the chat client. Basicaly streaming music. [22:15:10 CET] And its lagging, so its useless. I need to make it stop the lag [23:18:10 CET] Kwoth: use more buffering [23:18:39 CET] jasom: how? [23:19:39 CET] Kwoth: likely the skipping is happening in your C# library, so maintain a larger buffer there [23:20:00 CET] managed runtimes that aren't designed with bounded-work garbage collectors are going to be a bit of a pain for streaming audio [23:20:10 CET] i have no buffer in c#. [23:20:20 CET] i just pass the ffmpeg output to pcm sender [23:20:25 CET] in real time [23:20:27 CET] Kwoth: I would wager good money that this is your problem then [23:20:45 CET] Well, i know it is, i just want to know if ffmpeg can actually buffer it [23:20:58 CET] Kwoth: you need to buffer in C3 [23:20:59 CET] c# [23:21:13 CET] it doesn't matter how big the buffer in ffmpeg is if your c# library isn't reading from ffmpeg [23:21:40 CET] which pcm library are you using? [23:25:55 CET] what do you mean [23:26:01 CET] my c# library isnt reading from ffmpeg? [23:26:09 CET] i have process.start [23:26:15 CET] which runs this ffmpeg command [23:26:18 CET] and i read output stream [23:26:58 CET] right, and if your C# program is busy doing something else (e.g. a garbage collection) then it's not reading from ffmpeg [23:27:22 CET] nah, this is in a task [23:27:26 CET] it doesnt get blocked [23:27:48 CET] Kwoth: write the raw ffmpeg output to a file and see if it skips when you read from the file rather than a pipe [23:28:30 CET] kk, i will try that tommorow [23:28:42 CET] I did that before but i dont remember if it lagged or not [23:28:47 CET] Kwoth: if it allocates memory, it may get blocked. I don't konw CLR very well, but if it accesses memory on the heap, it may get blocked as well [23:29:39 CET] i just think the problem is it is not downloading fast enough. If i could prebuffer the part of the stream im reading from and predecode it, before i start reading from the output stream [23:29:43 CET] that would be perfect [23:31:48 CET] Ah, just found it; CLR uses a mark and compact GC that stops all threads during the GC [23:32:00 CET] so if any thread is allocating memory, your thread will get blocked from time to time [23:34:48 CET] i'm trying to make sense of a problem with some code based off of the filtering_audio.c example - i get the error "more samples than frame size" when trying to encode audio. As far as I can tell, the auto-inserted aresample filter is producing frames with a larger nb_samples than the encoder can handle. Do I need to call swr_convert manually? Or is there some way to influence the calculation in the auto-inserted resampler? [00:00:00 CET] --- Wed Dec 2 2015 From burek021 at gmail.com Wed Dec 2 02:05:02 2015 From: burek021 at gmail.com (burek) Date: Wed, 2 Dec 2015 02:05:02 +0100 (CET) Subject: [Ffmpeg-devel-irc] ffmpeg-devel.log.20151201 Message-ID: <20151202010502.F22CD18A0192@apolo.teamnet.rs> [00:03:04 CET] ffmpeg 03Michael Niedermayer 07master:40063a900b6d: ffserver: Clear avio context after closing it [00:03:04 CET] ffmpeg 03Michael Niedermayer 07master:a00cc2e40d40: ffserver: Clear avio context after closing it in rtp_new_av_stream() [00:03:04 CET] ffmpeg 03Will Kelleher 07master:0eac93da0fd5: qsvenc: write a53 caption data to SEI [01:06:33 CET] alright, I think all I need to finish up this patch series is a lavf version bump [01:10:01 CET] does adding fields to AVOutputFormat need a minor or a micro bump? [01:10:30 CET] also, adding a new public API to lavf [01:13:31 CET] minor [01:14:26 CET] jamrial: for both? [01:14:32 CET] (they're in different commits) [01:14:33 CET] yeah [01:14:34 CET] micro is for things like new avoptions in decoders/demuxers, or changes to existing ones [01:14:42 CET] alright [01:15:07 CET] if they are going to be commited at the same time, you can probably just bump once for both [01:16:00 CET] should that go in the second bump-needing commit, or in a separate one? [01:18:09 CET] in the second should be ok [01:24:55 CET] rcombs: you should also add a couple lines to apichanges [01:25:05 CET] oh good point [01:25:17 CET] and maybe to to Changelog as well [01:28:29 CET] if it's just a new API then apichanges is enough [01:28:39 CET] you're not adding a new component or feature [01:30:20 CET] I'd call it a feature, even if it's not exposed as one [01:38:28 CET] sometimes API changes are indeed in the changelog [01:38:38 CET] "not exposed"? [01:43:49 CET] wm4: like, it's not a new public API or component or anything [01:44:28 CET] it's new internal handling that makes some things easier [01:45:20 CET] but it makes live easier for the API user by not having to do the BSFing manually, right? [01:45:28 CET] yeah [01:45:36 CET] or the ffmpeg.c user [01:49:08 CET] well that makes it even more visible [01:49:17 CET] and should definitely be mentioned in the changelog [02:02:52 CET] ffmpeg 03Ganesh Ajjanagadde 07master:5a41a5a4f57d: avfilter/af_compand: use hypot() [02:02:53 CET] ffmpeg 03Ganesh Ajjanagadde 07master:7b11eead1b4e: avcodec/ac3: always use hardcoded tables [11:15:52 CET] do I see correctly that MPEG-2 Video doesn't have >8bit coding? It does have intra_dc_precision but that seems to be something else, right? [13:19:40 CET] JEEB: correct [13:27:16 CET] ffmpeg 03Michael Niedermayer 07master:8e7f4520226d: avformat/dump: Fix integer overflow in av_dump_format() [13:27:16 CET] ffmpeg 03Michael Niedermayer 07master:3a9cb18855d2: avutil/integer: Fix av_mod_i() with negative dividend [13:27:17 CET] ffmpeg 03Michael Niedermayer 07master:25e37f5ea92d: avutil/mathematics: Do not treat INT64_MIN as positive in av_rescale_rnd [14:46:39 CET] lol poor Ganesh is on rampage [14:50:51 CET] ... [14:55:51 CET] that "FFMPEG API" is going to be golden [14:55:56 CET] +thread [15:04:23 CET] QNAP is using ffmpeg for the Media Transcoding in their NAS! [15:04:34 CET] ffmpeg version 0.8.10, Copyright (c) 2000-2011 the FFmpeg developers, built on Nov 18 2015 03:42:44 with gcc 4.8.2 20131014 (prerelease) [15:05:05 CET] And they build it with --disable-yasm. [15:05:23 CET] something something idiots [15:05:24 CET] if they are remuxing probably not a big deal [15:05:29 CET] but yes if they are transcoding yes [15:05:32 CET] They are transcoding. [15:05:40 CET] from what to what? [15:05:46 CET] BtbN: is it x86 or something else? [15:05:50 CET] arm [15:06:02 CET] then --disable-yasm is just redundant [15:06:22 CET] wm4: cargo culted configure options, who would have ever heard of that [15:06:34 CET] kierank, basicaly everything you want. They built a Transcode-WebUI [15:06:53 CET] it's always fun when one of those uncountable combinations of configure flags actually cause problems [15:07:10 CET] It's actualy a quite short configure line. [15:07:22 CET] --prefix=/opt/cross-project/arm/linaro/arm-linux-gnueabihf/libc/usr --enable-cross-compile --cross-prefix=arm-linux-gnueabihf- --disable-static --enable-shared --disable-yasm --arch=arm --target-os=linux --disable-armvfp --enable-gpl --disable-encoder=snow --disable-decoder=snow [15:08:24 CET] Hm, enable-gpl though. [15:08:25 CET] why disable the best codec [15:08:58 CET] but I suppose they still want to encode to SVQ [15:09:04 CET] (really, why disable snow specifically) [15:13:17 CET] they must hate wavelets [15:15:36 CET] or they don't know what it is [15:16:06 CET] or fell in love with fourier 20 years ago and are afraid of getting further spatial information [15:16:10 CET] one does not know [15:20:33 CET] fritsch: btw. did you say you'd bisect the mp4 mp3 failure? [15:20:40 CET] or was that someone else [15:20:49 CET] wm4: i could - if it is wanted [15:21:02 CET] someone said, that we should talk with the handbrake dev before that [15:21:25 CET] yeah, I didn't follow this so well (I know there was some other change that fixed something else) [15:21:28 CET] i was the oppinion that if "quicktime cannot play it - one dos not have to care too much" [15:21:35 CET] yes, this one: [15:21:40 CET] that's probably true [15:21:51 CET] https://trac.ffmpeg.org/ticket/4938 [15:21:56 CET] fixes something similar [15:22:15 CET] would be funny if this commit would break the new file :-) [15:22:38 CET] depends what the default for parsing is [15:22:53 CET] bisecting should be easy to do [15:23:04 CET] as one can just use ffplay and directly see if it's working [15:23:06 CET] on master [15:23:11 CET] while start at ~ 2.7 [16:14:41 CET] nevcairiel: thanks for the review, I found a bigger issue that seems to cover all cases mozilla reported in a single patch [16:14:46 CET] nevcairiel: so & gj! [16:16:10 CET] can somebody make me permanent op on #ffmpeg? [16:18:59 CET] oh, also, mozilla apparently ships ffvp9: https://bugzilla.mozilla.org/show_bug.cgi?id=1210219 [16:19:01 CET] I didnt know that [16:19:21 CET] maybe I should give firefox a try again, chrome doesnt ship ffvp9 [16:20:50 CET] wm4: dies reverting the mentioned "fix" make it work again? [16:21:01 CET] fritsch: didn't try yet [16:29:38 CET] fucking threading [16:29:41 CET] i hate it so much [16:30:12 CET] atomic signaling which aren't shit atomic [16:32:15 CET] what's your problem? [16:32:28 CET] and what isn't atomic? [16:34:10 CET] see http://ffmpeg.org/pipermail/ffmpeg-devel/2015-December/184281.html [16:34:42 CET] two senders are waiting, the receiver is able to signal unlock and re-acquire the fucking mutex before any of the other wakes up [16:35:21 CET] well, pthread_cond_signal is basically an optimization to wake up only exactly 1 thread [16:35:30 CET] yes but not instantly [16:35:46 CET] apparently [16:35:51 CET] ? [16:36:10 CET] look at the log [16:36:47 CET] RECEIVER: acquired lock [16:36:47 CET] RECEIVER: reading a msg from the queue [16:36:47 CET] RECEIVER: signal the cond [16:36:47 CET] RECEIVER: acquired lock [16:36:49 CET] no unlock? [16:37:01 CET] i just didn't print it [16:37:04 CET] but it unlocks [16:37:12 CET] (after signal) [16:37:46 CET] def/win28 [16:37:52 CET] welp, sorry [16:38:12 CET] if senders can block, then a single signal is definitely wrong [16:38:26 CET] hm [16:38:28 CET] or is it [16:38:40 CET] it wouldn't be wrong if there were 2 conditions [16:38:59 CET] i'm going the simple way and broadcasting everyone [16:39:13 CET] (both recv & senders everytime) [16:39:25 CET] is it even supposed to have multiple readers and writers? [16:39:47 CET] now it is [16:40:04 CET] and i'm assuming yes [16:40:17 CET] since it's used like this in ffmpeg.c [16:40:23 CET] (1 thread per file) [16:40:40 CET] (used for packets queue) [16:40:52 CET] sounds fine then (but why does ffmpeg.c need that) [16:41:58 CET] -i in1 -i in2 -i in3 ... [16:42:04 CET] parallel demuxing [16:42:17 CET] shouldn't it have multiple queues then? [16:46:18 CET] dunno, didn't look closely [16:51:52 CET] BBB: i'm not sure i like the new patch, it makes integration with hwaccel harder, can we have a way that only sets ctx->pix_fmt in one place, ever? :) [16:52:57 CET] BBB: maybe write into s->pix_fmt and only set ctx->pix_fmt in update_size? [16:54:09 CET] what's up with the mozilla reports btw? the security ml just got flooded by them or sth? [17:09:17 CET] they started using ffmpeg over gst recently [17:09:18 CET] iirc [17:09:45 CET] ah, yeah right ok [17:09:54 CET] over gst? sadface [17:10:03 CET] i mean, instead of [17:10:08 CET] OH [17:10:11 CET] ok [17:31:13 CET] nevcairiel: ctx->pix_fmt is only set in read_colorspace_details(), so Im not sure theres an issue [17:32:59 CET] nevcairiel: I guess youre concerned because you want a single place where updates to ctx->pix_fmt are callbacked into hwaccel, right? [17:33:07 CET] nevcairiel: I think that doesnt change, its still update_size() [17:47:27 CET] Hello! [17:54:12 CET] I have been looking into converting equirectangular panoramic video into cubic video, but have not come across anyone who has created an FFMPEG filter for doing so. [17:54:34 CET] tha sounds incredibly niche [17:54:36 CET] so i am not surprised [17:54:38 CET] I would like to avoid having to write the filter from scratch, does anyone know of someone who has been working on something similar or a forum where I can ask around? [17:54:50 CET] Thats what I figured [17:56:11 CET] do you have samples of both videos [17:56:27 CET] Yes [17:56:29 CET] here's the right place to ask I'd say [17:56:55 CET] There are applications that convert Equirectangular videos to Cubic but it requires the conversion to happen manually and we would like to automate the process [17:57:47 CET] I can write it if I know how it works [17:58:04 CET] That would be amazing, I already found a Java implementation that works [17:58:13 CET] application name? [17:58:34 CET] Java is trivial to RE [18:00:11 CET] BBB: there is 4 lines that set ctx->pix_fmt [18:00:18 CET] What is RE, I thought ffmpeg filters were written in C? [18:00:20 CET] Mo_YouVisit : i remember reading about a filter to convert files into a 360 , like omnimax . lets see if i can find it [18:00:31 CET] BBB: in distinctly different places in the code [18:00:38 CET] The java code can be found here http://os.ivrpa.org/panosalado/downloads in EquirectangularToCubic.zip [18:00:56 CET] The conversion is doen in Equi2Rect.java [18:01:27 CET] nevcairiel: theyre all inside read_colorspace_details, or in the bug clause surrounding intraonly frames (where that syntax element is missing from the bitstream) [18:01:29 CET] nevcairiel: so ... [18:01:36 CET] nevcairiel: what is your practical worry? [18:01:37 CET] BBB: personally I would just use s->pix_fmt everywhere in vp9.c, and only sync that to ctx->pix_fmt when needed, ie. never actually read ctx->pix_fmt [18:01:49 CET] thats how most other decoders work [18:01:49 CET] but that reintroduces the bug [18:01:56 CET] how so [18:01:58 CET] Mo_YouVisit: than its even easier [18:02:04 CET] where s->bytesperpixel is not in sync with ctx->pix_fmt [18:02:16 CET] why does it matter, s-> is private, and if ctx->pix_fmt is never read [18:02:20 CET] if read_csp_details ssets s->bytesperpixel, it needs to set ctx->pix_fmt also [18:02:26 CET] i cant read the damn mozilla bug because they are paranoid tards [18:02:32 CET] theres nothing in it [18:02:34 CET] just a file [18:02:42 CET] durandal_1707: how so? [18:03:21 CET] BBB: in practice, any place that sets ctx->pix_fmt needs to call the hwaccel code and handle errors and whatnot [18:03:29 CET] I like to have that once [18:03:39 CET] not 4 times [18:03:49 CET] so lets do it in update_size()? [18:03:58 CET] what hwaccel code do we need to call? [18:04:08 CET] Mo_YouVisit: I will look at it when I go back home [18:04:55 CET] thats what I was suggesting, just don't ever read ctx->pix_fmt and only set it in update_size, instead use (a new) s->pix_fmt throughout the entire decoder, then this one doesn't desync, and no harm is done [18:05:02 CET] durandal_1707: Thank you! I will stay on IRC, just in case my email is mo.ben at youvisit.com [18:05:06 CET] nevcairiel: we cant do that [18:05:09 CET] why not [18:05:15 CET] nevcairiel: ctx->pix_fmt is used for buffer allocation [18:05:21 CET] so we need to set that, not a placeholder [18:05:32 CET] its functionally confirmed in update_size [18:05:36 CET] are buffers allocated before update_size? [18:05:44 CET] nope, but theres error paths before it [18:06:09 CET] then i dont see the problem [18:06:13 CET] so right now the error path allows us to a) set s->bytesperpixel, b) set s->pix_fmt (although its a local variable), then error out [18:06:22 CET] and never reach c) set ctx->pix_fmt [18:06:29 CET] this causes the two to be out of sync [18:06:40 CET] then the next (inter) frame continuous where internal state left off (being invalid) [18:06:44 CET] Mo_YouVisit: nevermind, cant find what i was looking for. [18:06:45 CET] continues [18:06:50 CET] and all kind of shit happens [18:06:54 CET] which doesnt really do harm, you didnt reach any code that would need it, and the next frame would call update_size and update ctx->pix_fmt if needed [18:06:55 CET] so I want them to always be synced [18:06:58 CET] Compn: Thats ok, thanks for the help! [18:07:13 CET] the next frame never calls update_size [18:07:22 CET] because the internal state is inconsistent [18:08:12 CET] you reviewed the original patch, you can see that we check for subsample+bpp changes through fmt [18:08:19 CET] if fmt never changes because we went into an error path [18:08:23 CET] the buffers are never reallocated [18:08:34 CET] but internal variables like s->bpp and so on are changed [18:08:37 CET] thats a huge issue [18:08:40 CET] i don't see how changing which variable you check against makes any kind of difference there [18:08:51 CET] you do the same check [18:09:06 CET] ok, lets start from zero [18:09:19 CET] Mo_YouVisit: do you have input samples? [18:09:20 CET] do you see how s->bytesperpixel and ctx->pix_fmt get inconsistent in the current code? [18:09:33 CET] if you trigger an error path between read_colorspace_details and update_size [18:09:41 CET] (of which theres a couple) [18:10:35 CET] regardless of whether that inconsistency is a problem or not, just, do you see theyll be inconsistent [18:10:38 CET] thats why I am suggesting you introduce s->pix_fmt which is always in sync with s->bpp, then you can use both these variables to check the internal state, and always keep everythign fine and dandy - and all that on top of your current patch which adds last_fmt to track changes [18:11:02 CET] just only have one central place as late as possible that updates ctx->pix_fmt [18:11:06 CET] but why [18:11:06 CET] like, in update_size [18:11:15 CET] I dont see why add an extra variable thats just a copy [18:11:19 CET] soooooo many variables [18:11:26 CET] because its not a copy for hwaccel [18:11:48 CET] s->pix_fmt would be yuv420p while ctx->pix_fmt would be pix_fmt_dxva2 [18:11:50 CET] or something [18:11:57 CET] that allows you to still do the proper checks [18:12:10 CET] aaaaaaha [18:12:10 CET] ok [18:12:14 CET] that starts to explain things [18:12:20 CET] so you dont ever want me to check ctx->pix_fmt, ever [18:12:24 CET] right [18:12:35 CET] never read that one, as hwaccels change it [18:12:36 CET] ctx->pix_fmt should be write-only [18:12:50 CET] now were getting somewhere [18:12:51 CET] ok [18:14:33 CET] it was a bit annoying in my current hwaccel patch as it is, but because you used a local variable i could work around it [18:14:45 CET] but by moving the setting into the reading function, it gets difficult [18:15:14 CET] ok, Ill change that [18:15:45 CET] how this filter should be named? [18:15:58 CET] re-testing with asan, will take an hour or so [18:16:07 CET] (compiling ffmpeg is sooooooooooooo sloooooooooooooow) [18:16:44 CET] there is still one weird thing left regarding pixfmts [18:16:54 CET] because vp9 is so crazy and supports ref frames in different sizes [18:17:23 CET] that doesnt affect pixfmt though [18:17:35 CET] no, but there is a check that checks if the ref is the same pixfmt [18:17:42 CET] which can be slightly difficult [18:17:46 CET] i'll have to re-visit that [18:18:09 CET] is the avframe->format using hwaccel or native pixfmt? [18:18:14 CET] hwaccel [18:18:17 CET] which is the problem [18:18:22 CET] hum&. right [18:18:27 CET] i'll think on that [18:18:32 CET] dont worry about it yet [18:18:43 CET] that would mean its currently broken right? [18:18:54 CET] i hacked around it in an ugly way [18:19:00 CET] ? [18:19:19 CET] I told it to consider a hwaccel foirmat equal to yuv420p [18:19:31 CET] which is generally true, but still ugly [18:19:55 CET] hahahahha [18:19:57 CET] so ugly [18:20:57 CET] actually [18:21:13 CET] hm no [18:21:51 CET] well i'll think on a solution [18:22:15 CET] there is also a problem with the location of that check [18:22:31 CET] it checks the format before update_size is called, so it doesnt even know if the hwaccel is used [18:22:43 CET] maybe I need to move that to another place to be sure [18:24:17 CET] will have to re-visit this [18:24:45 CET] better this way? [18:24:48 CET] (new patch on ML) [18:25:15 CET] nevcairiel: we can move that check down, its not critical where its placed, just that its done before read_header returns [18:25:23 CET] so moving it below update_size is called is fine [18:28:24 CET] BBB: patch looks fine, a quick grep shows ->pix_fmt isnt accessed anywhere else otherwise, so that seems ok [18:28:37 CET] right, I ctrl-fed that also :-p [19:04:20 CET] durandal_1707: I have an input uploaded here: http://video-us-east-1.youvisit.com/242477/2880_30_3.mp4 I guess it would make sense for it to be named EquirectangularToCubic [19:05:42 CET] Mo_YouVisit : did you see this https://github.com/eVRydayVR/ffmpeg-unwarpvr/ [19:06:12 CET] i guess thats just unwarping some edges for occulus rift [19:07:02 CET] Compn: Yes I have, that is just to remove barrel distortion like you said [19:10:50 CET] i thought there was another one [19:10:51 CET] for mplayer [19:11:10 CET] possibly using opengl for panoramic warping [19:11:21 CET] guh if i can find it [19:11:30 CET] not important i guess [19:15:06 CET] yes [19:15:13 CET] planetarium mplayer [19:15:23 CET] how does filter_complex deal with files having different bit depths? [19:15:32 CET] swscale? [19:15:57 CET] probably [19:16:02 CET] depends on what filters you use [19:16:35 CET] filters can announce what formats they want on their inputs [19:17:15 CET] ah thats just distorted fishe ye [19:17:58 CET] Yeah, there are a lot of distortion correction filters out there but not many for changing between projections [19:18:12 CET] fair enough, sorry for the noise : ) [19:18:23 CET] No problem, appreciate the help! [19:26:40 CET] whos working on filters and wants to merge that unwarp filter? https://github.com/eVRydayVR/ffmpeg-unwarpvr/ [19:26:45 CET] ubitux? [19:26:48 CET] :P [19:26:54 CET] Action: Compn runs [19:27:06 CET] wm4: starting to bisect now [19:29:31 CET] if the unfolded cube is this [19:29:34 CET] http://imgur.com/xtt2UZ3 [19:29:59 CET] with faces labeled as Left, Front, Right, Back, Up, Down [19:30:15 CET] we'd like the final video to be packaged like this: [19:30:22 CET] http://imgur.com/2PiDuh3 [19:33:27 CET] where the black bar is an 8 bit separation to avoid compression artifacts [19:34:01 CET] *8 pixel, not 8 bit [19:39:58 CET] interesting discovery today that dirac decoder has bugs [19:45:44 CET] are bbc still using dirac ? [19:46:26 CET] yes [19:46:28 CET] Action: kierank is too [19:46:36 CET] didn't you hear, wavelets are the next big thing in video encoding [19:47:00 CET] Action: atomnuker realizies it's not 2004 and wavelets suck for low-bitrate stuff [19:51:34 CET] Compn: i don't care [19:51:38 CET] wavelets :-) [19:51:39 CET] hehe [19:51:40 CET] but thx i guess [19:51:47 CET] the train is already driven further ... [19:51:56 CET] after it was quite popular at our institute ... [19:53:58 CET] atomnuker, bbc uses it as a mezzanine, as is tradition [19:54:04 CET] so no low bitrate [19:54:58 CET] I also think it's not really ture what @atomnuker said - at least not in general [19:55:02 CET] and not for all pairs [19:55:08 CET] (there are too many specialized) [19:55:40 CET] s/pairs/families/ [19:56:05 CET] it's true for all practical purposes currently [19:56:43 CET] makes me wonder a bit, cause already in 90s special wavelet for very low bitrate video has been suggested [19:57:29 CET] we also loved VQ and codebooks in the 90s ;) [19:57:30 CET] hybrids, using the wavelet for the intraframe pictures [19:57:34 CET] hehe [20:10:55 CET] zlib, zlib everywhere! [20:13:57 CET] well wavelets are still around, vp9 and daala use wavelets for lossless [20:14:29 CET] as they are in industrial image processing for defect classification [20:14:35 CET] daala also uses the haar wavelet for screencasting because it performs better on text than the DCT [20:14:41 CET] and if there is no argument left, you say: jpeg 2000 [20:15:23 CET] I've actually seen good jpeg 2000 encoders! [20:15:45 CET] but most, along with jpeg 2000 and the selection of the 9/7 wavelet, are based around hyperoptimizing PSNR [20:16:48 CET] I will just push my filters without waiting for ganesh approval, as he appears to dislike me in general [20:17:23 CET] which filters? [20:19:30 CET] TD-Linux: performs better means: better compression while psnr still better ,right? [20:19:51 CET] TD-Linux: which is kind of clear for the edges that this text has ... [20:22:48 CET] llogan: sidechaingate, apulsator, aemphasis [20:24:50 CET] fritsch, yeah, looks better at the same bitrate [20:24:51 CET] durandal_1707: i looked at apulsator docs. LGTM other than a few nits that I don't care about, but i'm sure ganesh will point them out. [20:25:39 CET] (i forgot to send apulsator LGTM email/noise earlier) [20:26:17 CET] also wavelets are used for DC coeffs in H264 and daala as well, in the former for speed, in the latter for flexibilty with varying block size [20:26:46 CET] yeah I wondered that the others sounded that negative about them [20:26:53 CET] really the only place they died was for the primary transform for natural images, where they give a worse coding gain. [20:27:39 CET] Timothy_Gu: did you ever look into a twitter bot to squawk releases? (I didn't) someone on twitter complained about my slowness/laziness. i don't tihnk it's important unless you are bored and want to look busy in class [20:28:09 CET] they were overhyped because they gave ridiculous PSNR numbers, and for features that no one wanted (multiresolution images), hence the hate :) [20:28:31 CET] TD-Linux: in germany the state "sponsored" new science project - if you had "wavelets" in the name ... [20:28:44 CET] so everyone started to be a wavelet specilist [20:29:10 CET] for classification: find the wavelet that displays the defect best ... while a simple highpass would have been sufficient [20:29:21 CET] or using the scales ... "just for fun" [20:29:25 CET] this is a scale 7 error [20:29:29 CET] this one a scale 4 [20:29:37 CET] and classifiing on the scale [21:23:05 CET] is checking for cabs and cexp enough for aemphasis? [21:35:30 CET] ok so if there's no more comments I gonna push aemphasis [21:40:30 CET] g 42 [21:41:30 CET] durandal_1707: several users have wanted to change color of background and waveforms for showwavespic. what do you think about that feature? [21:48:51 CET] ah, they can do something with colorchannelmixer [21:49:22 CET] but showwaves filter outputs into gray only [21:49:58 CET] could open bug report maybe saste wants to work on that [21:51:13 CET] i gave a shitty example using colorbalance [21:51:44 CET] ill make the feature request. maybe later today. [21:58:43 CET] ffmpeg 03Paul B Mahol 07master:cde75e3150ff: avfilter/vf_histogram: remove deprecated stuff [22:06:00 CET] you can use colorlevels too [22:12:06 CET] ffmpeg 03Michael Niedermayer 07master:736e2e2c3008: avfilter/vf_shuffleframes: Assert that the case of an uninitialized ret does not occur [22:12:27 CET] why is ff_aac_tableinit inlined in a .h? [22:16:50 CET] [13FFV1] 15dericed opened pull request #15: Definition updates (06master...06definition-updates) 02http://git.io/vRvIR [22:17:52 CET] hi! would a change to vf_scale introducing per-frame reevaluation of scaling factor, allowing to use timestamp/frame number as constant, be welcome? I need to implement "popping" of overlays. Got it working with large sendcmd files, but that's ugly. I see no other way. [22:50:37 CET] ubitux, liekly because it is shared in the enc and dec [22:56:17 CET] wm4: [6ec688e1bc76dd93151cbca1c340162ae4b10d77] mp3: enable packed main_data decoding in MP4 [22:57:15 CET] ok [22:57:23 CET] i will bisect to finish (1 step left) [22:57:38 CET] but I am quite sure that's exactly it [22:58:22 CET] you see the commit? i really wonder what this commit does [22:58:27 CET] especially at the beginning [22:58:44 CET] it removes some "framesize" checking [22:58:49 CET] and enables AVSTREAM_PARSE_FULL [22:59:14 CET] i will update the bugreport after bisecting is finishd [22:59:25 CET] seems like it rather removes AVSTREAM_PARSE_FULL, except in a special case [22:59:27 CET] but from an uneducated guess it seems the parse_full is not working [22:59:32 CET] jep [23:00:02 CET] but I really wonder why does it change the decoder and a single demuxer at the same time [23:01:46 CET] is luca barbato online here? [23:01:53 CET] "In order to accept this, this patch removes unnecessary frame size checking on mp3 decoder. [23:01:56 CET] " [23:02:05 CET] i always find it funny if it seems - that those were necessar [23:02:06 CET] y [23:02:40 CET] the real problem in that patch is limiting the parsing to a few select samples only [23:02:45 CET] yes [23:02:56 CET] that seems "adventurous" [23:03:01 CET] (in the right spelling of course) [23:04:04 CET] fritsch: he's lu_zero on Libav, but talking to him would be rude, because you're using and discussing FFmpeg, not Libav [23:04:52 CET] fritsch: query lu_zero and he will answer [23:05:40 CET] wm4: okay [23:05:45 CET] i don't want to be rude :-) [23:05:54 CET] i just reverted his patch, fixed the hunks [23:05:59 CET] and will now retest latest stable [23:06:08 CET] this bug has caused me headaches a very long time [23:06:12 CET] not only in this special mp4 [23:06:21 CET] but always from time to time with handbrake encodes [23:06:31 CET] I mean as long as you don't assume lu_zero is a ffmpeg dev it's probably ok [23:06:40 CET] ah okay [23:06:47 CET] i thought this libav / ffmpeg thing is over [23:06:47 CET] but then I'd also try to reproduce it with Libav [23:06:54 CET] haha, no [23:07:03 CET] (and reasonably speaking, ffmpeg could have broken it on merge) [23:07:09 CET] a let's see [23:07:17 CET] what cehoyos says [23:07:30 CET] better dont [23:07:37 CET] it'll just frustrate you [23:07:52 CET] hahaha [23:08:00 CET] hihi - i love people with principles :-) [23:08:08 CET] if that fritsch guy does not provide ffmpeg -i info [23:08:12 CET] i won't say thank you [23:08:14 CET] it's okay :-) [23:09:11 CET] Okay - let's play this game - i bet he checks it himself - as he is eager to close the bug [23:09:30 CET] while I think from a user pov I invested enough time ... [23:10:54 CET] michaelni: did you see the ffv1 segfault? [23:16:31 CET] fritsch: i think he just meant to point out that he already identified the regression in his comment [23:16:47 CET] nevcairiel: ah! [23:16:57 CET] man, if he would communicate more clearly [23:17:10 CET] i could have saved myself that bisect [23:17:40 CET] yeah - there it is [23:17:43 CET] but he was not sure .-) [23:17:55 CET] Regression since ae215e2b / 6ec688e1 [23:18:02 CET] 6ec688e1 <- was it in deed [23:18:13 CET] one is the commit, the other is the commit that merged it [23:21:01 CET] oki [23:21:03 CET] thx [23:22:02 CET] oki [23:22:07 CET] (including a typo) [23:23:10 CET] reverting should only be the last resort, usually there was a reason for the commit in the first place - fixing one sample at the expense of another is going to turn weird fast [23:24:14 CET] when reading the commit message [23:24:20 CET] i think it is even wrong [23:24:23 CET] in this case specifically, wm4 already send a patch always enabling parsing again in movdec, and apparently that breaks something [23:24:42 CET] let me test something [23:25:14 CET] yes, I got 2 failing samples, but didn't pursue it further [23:25:36 CET] (my patch might just be complete garbage) [23:26:27 CET] http://sprunge.us/hWUJ [23:26:33 CET] ^^ yours also does this, right? [23:26:52 CET] nah it removes another block of guessing [23:26:54 CET] there is two now [23:27:00 CET] okay [23:27:02 CET] but end result is the same [23:27:05 CET] parse full is set [23:27:29 CET] wm4: can you link me the non working sample? [23:27:34 CET] https://dl.dropboxusercontent.com/u/76042064/packed_maindata.mp3.mp4 [23:28:04 CET] yes, that [23:28:05 CET] yeah :-) [23:28:33 CET] btw. my old mpv version that played "the denis sample" also fails with this one [23:28:45 CET] gstreamer stutters [23:28:55 CET] makes tuut ... tuuut [23:29:00 CET] and quicktime? [23:29:10 CET] mpg123 fails to decode [23:29:20 CET] only linux here, sorry [23:29:37 CET] so - if ffmpeg cannot play _this_ fail it's not really a regression :-) [23:29:48 CET] but the denis file is played by all others / including older ffmpeg version [23:29:53 CET] if it works today and doesnt tomorrow, its a regression [23:29:54 CET] :D [23:30:04 CET] this worked? [23:30:20 CET] i would assume the linked file works wiith current git master [23:30:32 CET] meanwhile I'm trying to write a recursive function in shell (god, fuck shell) [23:30:36 CET] it does not work with ffmpeg 2.7.3 [23:30:42 CET] thats not git master [23:30:42 CET] :P [23:30:52 CET] wait I try [23:30:54 CET] with master [23:30:56 CET] two seconds [23:31:00 CET] was 2.7.3 before that change that broke it? [23:31:12 CET] that's the funny part: i cannot say exactly [23:31:19 CET] first I bisected between 2.7.0 and latest master [23:31:22 CET] and only got bad [23:31:29 CET] then I started earlier 2k14 [23:31:43 CET] wm4: so what doesn't suck? [23:31:48 CET] this packed_maindata.mp3.mp4 is broken with master, too [23:31:49 CET] :-) [23:31:54 CET] so not sure which regression you search [23:32:24 CET] 6ec688e1bc76dd is not in 2.7, it was commited in june, long after the 2.7 branch [23:32:33 CET] good [23:32:42 CET] durandal_1707: nothing, but shell sucks a _lot_ [23:33:04 CET] it's literally like trying to do precision mechanics with stone age tools [23:33:11 CET] nevcairiel: okay - so basically there is no ffmpeg versio nat all that can play your dropbox file? [23:33:18 CET] what you need shell for? [23:33:28 CET] can windows media player play it :-) [23:33:28 CET] durandal_1707: ffmpeg's build system [23:33:38 CET] can vlc play it? [23:33:51 CET] can anything? what can? [23:33:56 CET] wm4: don't you also have a computer? :-) [23:34:24 CET] yeah, but just enough time to waste my time on irc [23:34:57 CET] i tried mplayer, mpv, ffmpeg 2.7.3, ffmpeg-master, ffmpeg-master with the revert we talked, mpg123, totem [23:35:03 CET] no one made that play [23:35:17 CET] Start wmp... [23:35:33 CET] need a windozer for this [23:35:49 CET] but before you search regression you should find a version that actually played it (TM) [23:36:39 CET] old mplayer [23:36:50 CET] is there a non old mplayer? [23:36:54 CET] durandal_1707: that means you can play it? [23:37:07 CET] nah [23:37:26 CET] fritsch: using ffmpeg master it errors a lot on the console, but the audio isnt too broken [23:37:29 CET] vlc does not detect the stream [23:37:42 CET] nevcairiel: the same is true for the denis sample [23:37:58 CET] do you use ffplay to actually hear what comes out? [23:38:11 CET] nah i dont have sdl, i transcoded to flac :D [23:39:02 CET] http://thread.gmane.org/gmane.comp.video.ffmpeg.devel/204316 [23:39:04 CET] ^^ so [23:44:07 CET] packed_maindata.mp3.mp4 cannot be played in quicktime [23:45:16 CET] cbsrobot-: thx [23:45:37 CET] nevcairiel: i reverted the wrong commit from master, when retesting plain master - ffplay from master can play it [23:46:58 CET] good night [23:47:23 CET] durandal_1707: are you still interested in filter propositions ? [23:51:29 CET] cbsrobot-: which one? [23:54:50 CET] durandal_1707: implementing LEQ(A) LEQ(M) and/or ITU-R 468 audio meter [23:55:36 CET] hmm is there some documentation? [23:56:03 CET] lavfi have peak meter [23:56:08 CET] bacly https://en.wikipedia.org/wiki/ITU-R_468_noise_weighting [23:56:20 CET] it's a weighted meter [23:59:24 CET] could probably steal the K weighting from the ebur128 filter and modify the weights for the 468 curve [00:00:00 CET] --- Wed Dec 2 2015 From burek021 at gmail.com Thu Dec 3 02:05:01 2015 From: burek021 at gmail.com (burek) Date: Thu, 3 Dec 2015 02:05:01 +0100 (CET) Subject: [Ffmpeg-devel-irc] ffmpeg.log.20151202 Message-ID: <20151203010501.528CF2AD6009@apolo.teamnet.rs> [00:06:46 CET] how do I set an output codec again? reading the manual is a pain [00:08:41 CET] -codec:v libx264 [00:09:14 CET] see -codec in http://ffmpeg.org/ffmpeg.html#Main-options [00:10:04 CET] uhh it's not working... [00:10:20 CET] ffmpeg -i Rick\ Astley\ -\ Never\ Gonna\ Give\ You\ Up\ \[HQ\]-DLzxrzFCyOs.mp4 -codec:v apng rick.png [00:11:00 CET] hmm apparently saving it as .apng works [00:11:13 CET] altho I should probably scale this down lol [00:22:02 CET] Did we just get Rickrolled? [00:22:04 CET] so uhh how do I replace the first frame of an APNG? [00:22:44 CET] hmm I can probably just hex editor it... [00:22:57 CET] but that's a pain tbh [00:38:25 CET] output as frames, replace desired frames, remux to apng [01:04:39 CET] if I have an audio file encoded in a floating-point format, will the decoded array still be a uint8_t*? If so, I should cast it, right? [01:46:22 CET] hi. Is there a quick way to view the documentation for encoding parameters for -c:v utvideo ? [01:46:58 CET] ffmpeg -h encoder=utvideo [01:47:42 CET] thank you c_14 [01:49:00 CET] c_14: I get this when I try that. https://bpaste.net/show/b5160f67b7ba [01:49:56 CET] Looks like the utvideo encoder doesn't have any private options. [01:50:52 CET] that makes life simple. [02:23:59 CET] Does ffmpeg support TrueHD decoding? [02:29:22 CET] AndrewMock: apparently it does. ffmpeg -h decoders | grep -i truehd [02:29:37 CET] Thanks! [02:29:39 CET] I am trying to merge two audio files where one starts with 4 seconds of silence. As a test I have two files that are exactly the same but one (call this file A) has been silenced for the first 4 seconds. Whenever I merge this, file A has the 4 seconds of silence truncated from the beginning and it starts from the first point where sound is made. Driving me nuts! Any advice or direction? [02:50:03 CET] nedrocks: hallo [02:50:19 CET] waressearcher2: hi! [02:52:06 CET] nedrocks: wie geht's es dir ? [02:52:16 CET] waressearcher2: mir gehts gut [02:52:25 CET] waressearcher2: und du? [02:52:40 CET] nedrocks: es geht's [03:13:54 CET] hey guys, having an issue during installation. using this guide (https://trac.ffmpeg.org/wiki/CompilationGuide/Centos). My path is configured to /root/ and I require /usr/local/bin/ [03:15:02 CET] I would assume prefix="$HOME/ffmpeg_build. $HOME should be changed to /usr/local/bin/ [03:15:50 CET] I have tried updating the path and repeated the guide. [03:16:28 CET] ffmpeg is still showing as prefix="root/..... [03:16:38 CET] any ideas where I'm going wrong [03:17:22 CET] shown where? [03:17:45 CET] your prefix would be simply /usr/local and then a bunch of other paths will default to $prefix/bin, $prefix/lib, etc [03:20:27 CET] I will paste the output of #ffmpeg [03:20:46 CET] ffmpeg ffmpeg version git-2015-11-25-839eb1c Copyright (c) 2000-2015 the FFmpeg developers built with gcc 4.4.7 (GCC) 20120313 (Red Hat 4.4.7-16) configuration: --prefix=/root/ffmpeg_build --extra-cflags=-I/root/ffmpeg_build/include --extra-ldflags=-L/root/ffmpeg_build/lib --bindir=/root/bin --pkg-config-flags=--static --enable-gpl --enable-nonfree --enable-libfdk-aac --enable-libfreetype --enable-libmp3lame --enable-libopus --enable [03:21:17 CET] You never uninstalled the old version [03:21:30 CET] ah, yep [03:21:56 CET] shoot [03:22:16 CET] sometimes is just takes fresh eyes [03:22:18 CET] thanks [03:22:31 CET] I will uninstall and reinstall [03:31:08 CET] If I want to convert a planar array to a non-planar one with swr_convert, should I convert use decoded_frame->extended_data, or decoded_frame->extended_data[0]? [03:31:22 CET] will swr_convert know about the multiples channels? [03:36:00 CET] hey guys, noob here. I'm trying to use ffmpeg to concatenate a bunch of files into one video, having a little trouble. I can't predict whether the files will have the same aspect ratio/resolution, framerate, or anything like that - downloading them from an API and I have no control over that [03:36:09 CET] satt: hallo [03:36:38 CET] satt: use the concat filter [03:37:26 CET] hey waressearcherZ, c_14 [03:37:36 CET] here's the current command I'm using [03:37:38 CET] ffmpeg ${argsInputs[@]} -filter_complex "${argsFilterText[@]} concat=n=$count:v=1:a=1 [v] [a]" -map "[v]" -map "[a]" /$d/output.mp4 [03:38:48 CET] it's in a bash script, where ${argsInputs[@]} is an array of the inputs, and ${argsFilterText[@]} is something like this: [0:v] [0:a:0] [1:v] [1:a:0] [2:v] [2:a:0] [3:v] [3:a:0].... [03:39:10 CET] c_14: Is that what you mean by concat filter? [03:39:22 CET] yes, what's your problem with the current setup? [03:39:25 CET] Something not working? [03:39:51 CET] c_14: It works sometimes, for some sets of videos, but not others [03:39:59 CET] here's the error I had last [03:40:01 CET] Input link in1:v0 parameters (size 376x668, SAR 0:1) do not match the corresponding output link in0:v0 parameters (376x668, SAR 6863:6871) [03:40:24 CET] obviously the SAR don't match [03:41:12 CET] probably add something like scale=iw*sar:ih for each input [03:42:08 CET] I have the scale (width*height) I'd like the final file to be (smallest of all the input's resolutions) [03:42:20 CET] c_14: Where would I add that? [03:43:39 CET] You'd have to manually query the resolution of all the input files beforehand, then you can just set the scale directly to that pre-concat [03:44:02 CET] Though you'll probably want to scale to a height and then pad if the video aspect ratio doesn't fit [03:44:38 CET] actually [03:44:49 CET] you might want to scale to the width and then pad the height [03:46:50 CET] yeah, whatever's smaller should get padding I guess. [03:47:16 CET] I am querying the resolution of all input files already, I'm just not 100% sure what to do with that info, where it goes in the command [03:47:41 CET] create a filterchain for each video in which you scale,pad [03:48:13 CET] and put that before the concat filterchain [03:48:49 CET] before -filter_complex in another section, or just in the string in there? [03:49:28 CET] filter_complex defines a filtergraph, filterchains are ';' separated sections in a filtergraph [03:49:35 CET] so, yes [03:49:35 CET] sorry man, real new to this [03:49:39 CET] ah ok [03:50:38 CET] So something like '[0:v]scale=x:y,pad=x:y:a:b[v0];[1:v]scale=x:y,pad=x:y:a:b[v1];[v0][0:a:0][v1][1:a:0]concat=n=2:v=1:a=1[v][a]' [03:50:57 CET] Just you'll need to generate all those values in your script [03:51:57 CET] I've already been grabbing the scale using ffprobe, should be doable [03:53:08 CET] probably something like scale=${shortest_width}:-2,pad=${shortest_width}:${shortest_height} [03:53:29 CET] I think you don't need the positional parameters for pad since it defaults to centered [03:55:29 CET] c_14: nice thanks! I appreciate the help a lot. here's a gist of my script as of right now if you care to take a look [03:55:37 CET] https://gist.github.com/StoryStar/c6394039ed78fcc0a785 [03:57:14 CET] what does the -2 do in scale=${shortest_width}:-2 ? [03:57:39 CET] scales preserving aspect ration to the nearest even height [03:57:45 CET] you can probably use -1 since you're padding [03:57:58 CET] It's just that some codecs don't like heights that aren't divisible by 2 [03:58:10 CET] (and pixel formats) [03:58:25 CET] ah ok cool [03:58:40 CET] Also, just a remark but on line 29 you could just do 'for f in $d/*.mp4; do' and space yourself the if on line 33 [03:59:27 CET] analogue for 68, and I'd probably do a if [[ -f /$d/mylist.txt ]] instead of the if on 88 and move it outside of the for [03:59:41 CET] oh nice didn't know that. I'm usually an ios/android dev, new to bash as well [04:02:29 CET] s/space/spare/ [04:05:10 CET] I'm running this command to convert mp4 and ogg files to mp4, where "-t 6" is the duration of ./in.mp4 (approximately): ffmpeg -y -t 6 -i ./in.mp4 -i ./in.ogg -c:v copy -c:a aac -strict experimental -map 0:0 -map 1:0 -f mp4 -shortest ./out.mp4 [04:07:09 CET] ... nevermind, I got it working [04:24:03 CET] c_14: hmm doesn't seem to have solved it, similar error [04:24:04 CET] [Parsed_concat_18 @ 0x7fb9aa504ae0] Input link in1:v0 parameters (size 376x668, SAR 0:1) do not match the corresponding output link in0:v0 parameters (376x668, SAR 6863:6871) [04:24:14 CET] Failed to configure output pad on Parsed_concat_18 [04:24:24 CET] maybe I screwed up [04:24:55 CET] I do need some more coffee. here's the filter string [04:24:56 CET] -filter_complex [0:v]scale=376:-2,pad=376:668[v0]; [1:v]scale=376:-2,pad=376:668[v1]; [2:v]scale=376:-2,pad=376:668[v2]; [3:v]scale=376:-2,pad=376:668[v3]; [4:v]scale=376:-2,pad=376:668[v4]; [5:v]scale=376:-2,pad=376:668[v5]; [6:v]scale=376:-2,pad=376:668[v6]; [7:v]scale=376:-2,pad=376:668[v7]; [8:v]scale=376:-2,pad=376:668[v8]; [v0] [0:a:0] [v1] [1:a:0] [v2] [2:a:0] [v3] [3:a:0] [v4] [4:a:0] [v5] [5:a:0] [v6] [6:a:0] [ [04:24:56 CET] v7] [7:a:0] [v8] [8:a:0] concat=n=9:v=1:a=1 [v] [a] -map [v] -map [a] ///Users/matt/Documents/JACKTHEAUTORIPPER/testvids/proc/sunnymabrey/output.mp4 [04:31:30 CET] still having an issue here. ./configure --prefix="/usr/local/bin/ffmpeg_build" is showing as configuration: --prefix=/usr [04:31:39 CET] this is after a clean install [04:32:14 CET] Installation went great on another server. Not sure whats happening here [04:32:52 CET] followed the guide found here: https://trac.ffmpeg.org/wiki/CompilationGuide/Centos [04:33:26 CET] hard coded $HOME as /usr/local/bin/ [04:33:30 CET] what are good settings for hls transcoding when on a small server? [04:34:04 CET] -hls_time 2 -hls_list_size 20 -hls_wrap 40 [04:34:20 CET] i'm using these numbers but i really have no idea. my stream is choppy as hell. [04:34:20 CET] on my working server, configuration: --prefix=/usr/local/cpffmpeg [04:34:47 CET] any recommendations on updating the prefix correctly here [04:35:05 CET] i usually do `--prefix=~/bin` [04:35:21 CET] or --prefix=/home/me/bin [04:35:32 CET] then i add ~/bin to my path up front [04:36:23 CET] I will give that a shot [04:36:39 CET] that's only for stuff i compile myself. [04:36:40 CET] echo ~ [04:36:50 CET] but i just found a build so that was easy [04:36:52 CET] displays /root [04:36:52 CET] you probably want --prefix=/usr/local [04:37:04 CET] yeah, thats what I'm trying to get [04:37:13 CET] i don't put manually compiled stuff in /usr/local [04:37:16 CET] don't set /usr/local/bin as the prefix [04:38:22 CET] so: ./configure --prefix="~/ffmpeg_build" --disable-examples [04:38:43 CET] I've tried: ./configure --prefix="$HOME/ffmpeg_build" --disable-examples [04:38:54 CET] and: ./configure --prefix="$PATH/ffmpeg_build" --disable-examples [04:39:45 CET] off to grab coffee, back in a bit [04:39:52 CET] definitely don't use $PATH [04:40:23 CET] yeah I cleaned that up [04:41:07 CET] I can't get the prefix set anything except /root/ [04:41:31 CET] unless I attempt to hard code it [04:42:07 CET] and thats not giving me the result I want, just trying to match the working server (sandbox) [04:42:11 CET] :configuration: --prefix=/usr/local/cpffmpeg [04:43:10 CET] did you run ./configure as root [04:43:19 CET] yeah [04:43:25 CET] logged in as root [04:43:35 CET] didnt sudo [04:43:59 CET] you can type `sudo bash` to avoid logging in as root [04:44:07 CET] are you now running it as a normal user [04:44:22 CET] if so then it's probably failing to overwrite autoconf files that you generated as root [04:46:06 CET] still root [04:47:18 CET] what's the least painful way to use FFmpeg in Java that isn't Xuggler? [04:47:29 CET] I'm going to try ~ instead of $HOME [04:47:42 CET] can't hurt at this point [04:58:14 CET] Rin_: if you're logged in as root then $HOME is /root [04:58:37 CET] also there's no reason to be root for anything but make install [04:58:47 CET] and not even that if you're installing to $HOME [05:04:03 CET] ok, I will logout and login as user and try again [05:04:04 CET] thanks [05:04:47 CET] you'll probably need to change the owner of the source now [05:04:58 CET] or just delete the directory and extract it again [05:09:54 CET] yeah I was thinking of deleting and starting over [05:11:37 CET] also yeah you should probably just install to /usr/local [05:13:36 CET] ok thanks, will do [05:13:44 CET] unless you specifically want a copy that's local to your user [05:29:23 CET] ffmpeg won't stream hls [05:29:29 CET] it starts then just goes berzerker [06:43:30 CET] hello everybody, im trying this ffmpeg -ss 00:00:01 -i Am_Watching_You_medium.webm -i palette.png -vf "drawbox=ih/PHI:color=black at 0.4:width=iw:height=48:t=max,drawtext=fontfile=arial.ttf:text='Im watching you Bitch':fontcolor=orange:fontsize=24:x=(w-tw)/2:y=(h/PHI)+th,paletteuse" -to 2 -y output.gif but i get an error about the filter that is complex ro something like that, can somebody give me a hint pl [06:43:36 CET] ease? [06:44:25 CET] im pretty new to ffmpeg and i dont know the order i should use on the commands and filters [06:45:09 CET] imple filtergraph 'drawbox=ih/PHI:color=black at 0.4:width=iw:height=48:t=max,drawtext=fontfile=arial.ttf:text='Im watching you Bitch':fontcolor=orange:fontsize=24:x=(w-tw)/2:y=(h/PHI)+th,paletteuse' was expected to have exactly 1 input and 1 output. However, it had >1 input(s) and 1 output(s). Please adjust, or use a complex filtergraph (-filter_complex) instead. <---- this is the error i get [06:48:02 CET] maybe this is easier to read http://pastie.org/private/fubtu8sm4ai3pstsxzkeq [06:52:06 CET] autofsckk: hallo [06:54:13 CET] hello waressearcher2 [06:55:22 CET] autofsckk: wie geth's ? [06:55:50 CET] i dont understand :/ [07:40:26 CET] Anyone around? [07:41:11 CET] Struggling with concatenating some videos [07:41:24 CET] Here's the error [07:41:25 CET] Input link in4:v0 parameters (size 376x668, SAR 0:1) do not match the corresponding output link in0:v0 parameters (376x668, SAR 6863:6871) [07:52:36 CET] satt: looks like the SAR is different for the two streams. you may be able to setsar them to the same value, or you may have to reencode one/both to concatenate them [07:53:40 CET] @pzich - thanks, yeah I thought so. Where would I setsar? I'm using this command [07:54:02 CET] ffmpeg (inputs) -filter_complex [0:v]scale=376:-2,pad=376:668[v0]; [1:v]scale=376:-2,pad=376:668[v1]; [2:v]scale=376:-2,pad=376:668[v2]; [3:v]scale=376:-2,pad=376:668[v3]; [4:v]scale=376:-2,pad=376:668[v4]; [5:v]scale=376:-2,pad=376:668[v5]; [v0] [0:a:0] [v1] [1:a:0] [v2] [2:a:0] [v3] [3:a:0] [v4] [4:a:0] [v5] [5:a:0] concat=n=6:v=1:a=1 [v] [a] -map [v] -map [a] [07:54:43 CET] I'm real new to ffmpeg, appreciate the help! [07:59:38 CET] possibly at the end of each of your scale and pad lines, I might try setsar=sar=1/1 or setsar=sar=0/1 [08:01:24 CET] pzich: thanks, I'll give that a shot right now [08:06:17 CET] pzich: Thank you!! It seems to have fixed it, for the 10-15 videos I just tested it on [08:06:30 CET] using setset=sar=1/1 [08:10:11 CET] no problem! [08:11:21 CET] hey question... i am using ffmpeg with nvenc for encoding... i am having trouble with decoding as the cpus seem to be weak.. source is h264 .. any way to decode h264 to mpegts using the GPU ? [08:45:32 CET] prelude2004c: there are hwaccels for everything in ffmpeg [08:45:40 CET] prelude2004c: not sure if that question is sane? [09:14:48 CET] i dont even know what i am asking :P haha [09:15:17 CET] basically the issue is this... i have a source h264 which i want to transcode into differnet streams and for that i use nvenc which uses up the GPUs.. and that works well. [09:15:41 CET] the issue right now is ... on the input side.. i am getting errors on decoding ... i think my cpu cores or something is limiting my decoding power [09:15:58 CET] so i wanted to find a way since our cards have both ENcoder and decoder and the decoder is at 0% usage [09:16:16 CET] so wondering if some way to use the GPU side of the cuda cores for decoding and then the hardware nvenc encoding on the cards for the encoding part [09:16:23 CET] this could solve my issue [09:16:30 CET] any ideas would ge breat [09:16:32 CET] great* [09:17:30 CET] anYc: had a chance to look at the avcut issue? [09:29:17 CET] bencc: sorry, not yet. been a bit busy with work lately but it's almost on top of the todo list :) [09:45:11 CET] Any ideas with decoding with gpu? [09:45:42 CET] Am I on the right track looking at hardware acceleration? [09:46:24 CET] prelude2004c_ZZZ: hallo [09:51:34 CET] anYc: great. thanks :) [09:54:10 CET] waressearcher2: stop! [09:55:43 CET] durandal_1707: nein [09:56:26 CET] Hi! I'm trying to write a C application for streaming video(frames being generated in real-time) using FFmpeg libav*. I've tried using the doc/examples/muxing.c as a reference and got it work but I want to broadcast the stream rather than a P2P connection. Does anyone have any pointers? [09:58:07 CET] durandal_1707: kann ich kamikaze_ "hallo" sagen ? [10:08:28 CET] kamikaze_: hallo, wie geht's ? [10:09:52 CET] @waressearcher2 Mir geht's gut [10:53:12 CET] hi. i am trying to encode a h264 file from a device to file and play it in a html5 video element at the same time. the recording goes fine, the playing works too, but the browser wants to load the entire file before it can play. i already use -movflags frag_keyframe+empty_moov but for some reason this doesn't seem to work. anyone has an idea? [10:54:38 CET] faststart [10:55:14 CET] durandal_1707, yeah, but that only works when the file is done recording [10:56:54 CET] dash? [10:57:22 CET] what? [10:57:52 CET] oh you mean streaming [10:59:37 CET] well i dont really need it to be like a constantly streaming endpoint [11:00:27 CET] it[ a logger system so a user can select an hour and a day and the video element then retrieves a video file [11:01:36 CET] and i could be able to do the faststart on the hours that are finished... but the hour that is currently recording is the problem [11:34:18 CET] Im trying to build from source on my Mac (for this bugreport of mine) but when I do ./configure --enable-libvorbis --enable-libvpx I get ERROR: libvorbis not found [11:36:21 CET] hmbg, i guess because you have to build vorbis and include it [11:37:03 CET] or at least grab a homebrew version of it :) [11:37:13 CET] I did install it with homebrew, no dice [11:37:23 CET] missing include path then? [11:37:30 CET] it's somewhere in /usr/local/Cellar or whatsit [11:37:34 CET] perhaps [11:37:39 CET] where do I add include paths? [11:37:56 CET] it's a configure parameter, don't remember OTOh [11:38:13 CET] or perhaps just --extra-cflags="-I@ [11:43:21 CET] trying ./configure --enable-libvorbis --enable-libvpx --incdir=/usr/local/Cellar/libvorbis/1.3.5/include/vorbis/, no dice [11:44:27 CET] look in config.log what's really going wrong [11:47:55 CET] #include fails [11:51:53 CET] I tried different lengths of the incdir, /usr/local/Cellar/ and up [11:51:56 CET] same error [11:52:00 CET] the file is most definitely there [12:16:11 CET] --extra-cflags=�-I�, like Mavrik said, should work [13:52:37 CET] I'm calling avformat_alloc_output_context2(c, NULL, "mpegts", NULL) and get -22, I'd say from the docs that the last argument can't be null if the second is null, but I fail to see why I'd need the second arg, -22 ain't related to this? [13:55:40 CET] the filename is null as the idea is to write to a socket not a regular file, if that doesn't matter I can put a dummy filename [13:58:19 CET] I am trying to capture video from tv tuner card but i get this error "[alsa @ 0x14df3d0] cannot open audio device hw:1,0" [14:15:46 CET] oh and btw I compiled with --enable-muxer=mpegts do I need some extra flag to get mpegts muxing? [18:19:42 CET] is there a vai so set a filename for ffmpeg log to log? I set av_log_set_level to both verbose and debug before avformat_alloc_output_context2 but it didn't output anything I suspect stderr or sth like that is closed or not attached to my console [18:36:53 CET] dantti: You'll probably have to set your own log callback with av_log_set_callback [18:37:26 CET] thanks [18:50:46 CET] now I wonder how "mpegts" turned into "Unable to find a suitable output format for 'xm�'" :P [18:51:03 CET] hmm no [18:55:28 CET] any idea why avformat_alloc_output_context2(&m_formatContext, NULL, "mpegts", "mpegts"); would become Requested output format 'x6l�' is not a suitable output format [18:55:59 CET] I get the same output when the last arg is NULL [19:00:18 CET] ok, my mistake on the log_callback [19:00:54 CET] what might be wrong then? "Requested output format 'mpegts' is not a suitable output format" [19:28:05 CET] was calling avcodec_register_all while I needed av_register_all... [20:02:02 CET] is there some AVIOContext that can give me a buffer so I write it to a socket? [20:11:08 CET] dantti, a custom one :) [20:11:23 CET] I mean, AVIOContext gives you a buffer by definition :) [20:11:43 CET] Mavrik: yeah, I was seeing some people do that on gg :P [20:11:57 CET] Mavrik: is there a way to be sure seek won't happen? [20:12:18 CET] you can set it non seekable, but then the process will crash :) [20:12:38 CET] There's a flat on the AVIOContext for that [20:12:51 CET] is there a way not to crash? would it crash even on mpegts? [20:13:02 CET] as I can't seek on a socket :P [20:13:49 CET] AVFMT_NO_BYTE_SEEK? [20:14:05 CET] mpegts won't seek [20:14:11 CET] it's meant as a streamable format :) [20:14:31 CET] ok, that's what I thought but better be double sure :) [20:14:38 CET] thanks [20:41:36 CET] hi. I have several videos from my sony camera here. MTS format. ffmpeg can read it, so thats good. But cinelerra can't. I now want to convert it to a different codec before editing. Something lossless compressed would be the right choice I guess. [20:41:38 CET] any recommendations? [20:42:01 CET] (I am a video editing beginner. This is the first real video I will cut) [20:46:12 CET] nilsge, what does ffprobe say on the MTS? [20:51:34 CET] Mavrik: this is the actual file, this was already processed by ffpmeg once (combining three smaller files to a bigger one) [20:51:36 CET] http://pastebin.com/GAs9r9y0 [20:51:51 CET] If I wanted to see all options available for libx264, would ffmpeg.exe -h -c:v libx264 be the proper syntax? [20:52:03 CET] louistwo: -h encoder=libx264 [20:52:06 CET] Mavrik: http://pastebin.com/vaF0aHBC this is a file directly from the camera [20:52:10 CET] Btw, are you sure you wanna use cinelerra? :) [20:52:12 CET] doh, I forgot from yesterday c_14 . thanks. [20:52:46 CET] Mavrik: no. Everything that is not based on MLT is fine [20:52:55 CET] which is pitivi and cinelerra, afaik [20:53:09 CET] I don't do video editing, but last time I've sen cinelerra it was kinda horrible :) [20:53:14 CET] louistwo: it won't display the stuff you can use with -x264opts/-x264-params, but 99% of users don't need to use those anyway (and probably shouldn't) [20:53:18 CET] nilsge, anyway, it seems that cinelerra just doesn't like your container [20:53:29 CET] i was urious to find what I can do with dnxhd as well [20:53:31 CET] nilsge, so try a simple ffmpeg -i .mts -codec copy output.mp4 [20:53:43 CET] i just wanted to make sure I got it right on a codec I knew would give me options since it is well supported by ffmpeg [20:53:48 CET] Mavrik: mp4 is lossless? [20:54:13 CET] or wait. this is not the codec, just the container, right? [20:54:23 CET] Dyep [20:54:27 CET] <- noob [20:54:39 CET] "-codec copy" means to just copy the streams (the H.264 video and aac audio) [20:54:44 CET] you see listed in ffprobe output ;) [20:54:58 CET] yeah, I read that :) [20:55:20 CET] I'm a little onfused on choosing a valid dnxhd profile [20:55:24 CET] it lists them but without a name [20:55:38 CET] I know audio codecs quite well, so I can deduce some stuff about video codecs. Also watched the primer from Xiph. [20:58:26 CET] hrmm? [20:58:38 CET] Action: Primer feels violated [20:58:43 CET] ha :) [20:58:43 CET] :-) [20:59:53 CET] Primer: don't feel bad. retard probably gets a few errant pings now and then [21:00:28 CET] Mavrik: I'm getting quite a few of these: [mp4 @ 0x559ec2d4d360] pts has no value97kB time=00:29:39.44 bitrate=21342.6kbits/s [21:00:41 CET] anything critical or safe to ignore? [21:03:46 CET] That's not nice [21:12:54 CET] If my USB microscope is being detected as a webcam, ffmpeg should be able to record a video stream, right? [21:13:28 CET] most likely [21:14:34 CET] Does it automatically detect the input device, or do I have to find it using some sort of -li command? [21:16:38 CET] I kind of want to make a timelapse using the USB microscope and have it encoded in x264 [21:29:57 CET] YaMoonSun: see http://ffmpeg.org/ffmpeg-devices.html#video4linux2_002c-v4l2 [21:30:12 CET] (assuming linux) [21:30:27 CET] also https://trac.ffmpeg.org/wiki/Capture/Webcam [21:31:30 CET] Cheers, windows at the moment, I have this link here https://trac.ffmpeg.org/wiki/Capture/Webcam [21:32:22 CET] llogan: I bet I'll need to edit the frame rate to 1 frame every 20 minutes or so to make a decent time laps [21:33:33 CET] furq, are you using the latest version? apparently the bug was fixed [21:59:50 CET] ./configure --disable-debug --enable-pic --optflags=-Ofast --enable-shared --disable-static --prefix=/usr --arch=x86 --disable-postproc --disable-swscale-alpha --disable-ffmpeg --disable-ffprobe --disable-ffserver --disable-doc --disable-htmlpages --disable-manpages --disable-podpages --disable-txtpages --disable-encoders --disable-decoders --disable-hwaccels --disable-muxers --disable-demuxers --disable-bsfs [21:59:50 CET] --disable-indevs --disable-outdevs --disable-devices --disable-filters --disable-parsers --disable-protocols --enable-decoder=h264 --enable-decoder=mjpeg --enable-parser=h264 --enable-parser=mjpeg --enable-demuxer=rtsp --enable-protocol=http --enable-protocol=httpproxy --enable-protocol=tcp --enable-protocol=udp --enable-filter=scale [22:00:23 CET] why does !CONFIG_FFPLAY=yes get inserted into the config.mak [22:00:28 CET] when doing that line of code [22:00:40 CET] It doesnt do that crap on x86 but on arm it comments out ffplay [22:16:21 CET] hi all [22:16:31 CET] i'm trying to find a reference to something i came across in a CTF [22:17:34 CET] apparently converting an HLS playlist (m3u8) that has a reference to a .txt file will instruct ffmpeg to conver the text file contents to a video frame [22:17:41 CET] see example playlist here: http://pastebin.com/raw.php?i=WFVuQN1v [22:18:07 CET] I'm trying to understand whether this is an FFMPEG specific feature or an HLS playlist feature but I can't find anything pointing me to that information [22:18:19 CET] I was wondering if anybody here could provide insight [22:19:07 CET] the ffmpeg command that was used was ffmpeg -i 'playlist' -f webm -codec:v vp8 -cpu-used 4 video.webm [22:19:25 CET] this apparently only works with .txt extensions [22:35:09 CET] hey, I'm trying to modify an existing application that uses ffmpeg, and I'm trying to use libavfilter.. but any filter I try to get with avfilter_get_by_name returns nothing: http://paste.scsys.co.uk/502322 [22:35:27 CET] that just prints "Error couldnt get filter\n" [22:35:35 CET] every time [22:37:39 CET] skaufman: you should call avfilter_register_all(); [22:38:03 CET] then avfilter_get_by_name will return something [22:39:05 CET] and so it does :) [22:39:08 CET] mateo`: thanks! [22:41:59 CET] hey guys.. good day.. can someone help me.. i am having trouble decoding h264 content.. i keep getting errors .. is there some way using nvidia GPU for me to use the card to decode instead of the CPU's ? [23:21:11 CET] prelude2004c: what errors? how can we duplicate the issue? [00:00:00 CET] --- Thu Dec 3 2015 From burek021 at gmail.com Thu Dec 3 02:05:02 2015 From: burek021 at gmail.com (burek) Date: Thu, 3 Dec 2015 02:05:02 +0100 (CET) Subject: [Ffmpeg-devel-irc] ffmpeg-devel.log.20151202 Message-ID: <20151203010502.8D0242AD6037@apolo.teamnet.rs> [00:00:01 CET] although that probably doesnt make it much easier than to just start over [01:06:36 CET] ffmpeg 03Ganesh Ajjanagadde 07master:789ef087c636: avcodec/ppc/fdctdsp: use more accurate constants [01:06:37 CET] ffmpeg 03Ganesh Ajjanagadde 07master:538e8ab512ab: avutil/tablegen: add tablegen libm compatibility shims [01:06:38 CET] ffmpeg 03Ganesh Ajjanagadde 07master:16d4c28cbb9a: avcodec/mpegaudio_tablegen: speed up dynamic table creation [01:06:39 CET] ffmpeg 03Ganesh Ajjanagadde 07master:1d0c94ddaef8: avcodec/mpegaudio_tablegen: more dynamic table creation speedups [01:06:40 CET] ffmpeg 03Ganesh Ajjanagadde 07master:2f5075f5513f: avcodec/cbrt_tablegen: speed up dynamic table creation [01:26:59 CET] ffmpeg 03Andrey Utkin 07master:648b26acc5e2: doc/filters/drawtext: fix centering example [01:33:19 CET] ffmpeg 03Ganesh Ajjanagadde 07master:8c2438053786: avfilter/af_volumedetect: use log10 instead of hardcoded constant [01:54:39 CET] [13FFV1] 15michaelni closed pull request #15: Definition updates (06master...06definition-updates) 02http://git.io/vRvIR [02:05:23 CET] Doesn't that enum ABI change being proposed need a major version bump? [02:32:16 CET] ffmpeg 03Andrey Utkin 07release/2.8:482bece29bc3: doc/filters/drawtext: fix centering example [02:32:18 CET] no [02:32:51 CET] ffmpeg 03Andrey Utkin 07release/2.7:70373bde2d5c: doc/filters/drawtext: fix centering example [02:33:02 CET] or rather, it needs to be wrapped in an FF_API define for the enum value to change once major is bumped [02:33:04 CET] ffmpeg 03Andrey Utkin 07release/2.6:95b70b656714: doc/filters/drawtext: fix centering example [02:33:17 CET] ffmpeg 03Andrey Utkin 07release/2.5:dc5bec14ab7d: doc/filters/drawtext: fix centering example [02:33:28 CET] ffmpeg 03Andrey Utkin 07release/2.4:a38a41df3989: doc/filters/drawtext: fix centering example [02:33:38 CET] ffmpeg 03Andrey Utkin 07release/2.3:f450caf05752: doc/filters/drawtext: fix centering example [02:33:49 CET] ffmpeg 03Andrey Utkin 07release/2.2:ed4c595edd73: doc/filters/drawtext: fix centering example [02:34:15 CET] ffmpeg 03Andrey Utkin 07release/2.0:c513e721f3fb: doc/filters/drawtext: fix centering example [02:36:30 CET] ffmpeg 03Andrey Utkin 07release/2.1:32ba667dc7fd: doc/filters/drawtext: fix centering example [04:38:02 CET] ffmpeg 03Michael Niedermayer 07master:e1057babdfb0: avcodec/mpegaudio_tablegen: Add (), fix fate [05:00:06 CET] hello! did anyone here work on the blu-ray subtitle format .sup decoder? https://ffmpeg.org/doxygen/2.8/pgssubdec_8c_source.html if so, can someone please link me the specs of the format? I need to render pgs subs over video in a certain app but I can't find info on it anywhere [06:07:42 CET] compiling C++ files with clang on Linux is currently broken [06:29:54 CET] hello! did anyone here work on the blu-ray subtitle format .sup decoder? https://ffmpeg.org/doxygen/2.8/pgssubdec_8c_source.html if so, can someone please link me the specs of the format? I need to render pgs subs over video in a certain app but I can't find info on it literally anywhere [10:38:58 CET] michaelni: how was mpegaudio_tablegen broken in 1d0c94dda? is the benchmark still the same? [10:40:01 CET] yvear: mmh i think we have a ticket about this [10:40:41 CET] yvear: what about libavformat/supdec.c? [10:41:05 CET] sup is a simple format muxing pgs subs [10:41:20 CET] basically a saner equivalent to the "vobsub" from dvds [10:41:40 CET] in general, many specs are usually not publicly available, you could read the ffmpeg .sup demuxer and the ffmpeg pgs decoder for a basic understanding [10:44:39 CET] isn't sup just an invention from third parties? [10:44:58 CET] like, a random guy in a forum distributing a extractsup.exe one day [10:46:04 CET] ubitux: got that impression too [10:46:29 CET] and it's also trivial [11:05:13 CET] ffmpeg 03Chris Cunningham 07master:5e6ce28dabe0: avformat/mp3dec: prefer "fast_seek" to TOC seek for CBR files. [11:09:16 CET] ubitux: i think so, yes, its the blu-ray subtitle format, why would any official body invent a standalone format :) [11:10:40 CET] ultraviolet or some shit, I guess [11:10:44 CET] ffmpeg 03Aaron Colwell 07master:ec83efd4d3c5: avformat/matroskadec: Fix sample_aspect_ratio for stereo matroska content [11:15:31 CET] switch and case must be below each other [11:16:35 CET] if you have a config to instruct vim to do so while reindenting, i'll love a share [11:47:57 CET] ffmpeg 03Claudio Freire 07master:ca203e9985cd: AAC encoder: improve SF range utilization [11:52:38 CET] new atomic bomb landing? [12:03:59 CET] ffmpeg 03Michael Niedermayer 07release/2.5:b622d6f6f65d: avcodec/ffv1dec: Print an error if the quant table count is invalid [12:04:00 CET] ffmpeg 03Michael Niedermayer 07release/2.5:fa3df7ae415b: avcodec/ffv1dec: Clear quant_table_count if its invalid [12:04:01 CET] ffmpeg 03Michael Niedermayer 07release/2.6:df660d685435: avcodec/ffv1dec: Print an error if the quant table count is invalid [12:04:02 CET] ffmpeg 03Michael Niedermayer 07release/2.6:a1b8fa1d9add: avcodec/ffv1dec: Clear quant_table_count if its invalid [12:04:03 CET] ffmpeg 03Michael Niedermayer 07release/2.7:183f58059442: avcodec/ffv1dec: Print an error if the quant table count is invalid [12:04:04 CET] ffmpeg 03Michael Niedermayer 07release/2.7:e6264f00addc: avcodec/ffv1dec: Clear quant_table_count if its invalid [12:04:05 CET] ffmpeg 03Michael Niedermayer 07release/2.8:1c6243228c34: avcodec/ffv1dec: Print an error if the quant table count is invalid [12:04:06 CET] ffmpeg 03Michael Niedermayer 07release/2.8:ff3e717003ef: avcodec/ffv1dec: Clear quant_table_count if its invalid [12:09:16 CET] ffmpeg 03Paul B Mahol 07master:c4f7b8f0db6e: avfilter: add audio pulsator filter [12:09:17 CET] ffmpeg 03Paul B Mahol 07master:6670527d33bb: avfilter/af_stereotools: remove floor usage [12:18:52 CET] can has experimental removal for aacenc now? [12:22:53 CET] klaussfreire: ^^ [12:23:08 CET] I'm all for it [12:39:43 CET] michaelni: thanks [13:35:55 CET] looking at the mpeg4 decoder, which is partially in h263dec.c [13:36:05 CET] MpegEncContext *s = avctx->priv_data; [13:36:06 CET] ok... [13:36:13 CET] * FIXME: By the way H263 decoder is evolving it should have [13:36:13 CET] * an H263EncContext */ [13:36:15 CET] ...ok. [13:37:12 CET] that comment is from 2001 [13:37:48 CET] of course it is [13:38:35 CET] :) [13:38:53 CET] spot the number of things that are wrong with the code quoted above [13:43:18 CET] lol [14:10:08 CET] wouldn't it be better if mpeg4 repeated a frame with empty VOP elements (vop_coded=0) instead of just skipping them? ( [14:14:33 CET] Why [14:14:56 CET] because there are evil AVIs which have such packets to do some sort of pseudo-VFR [14:15:19 CET] Action: kierank blames VFR [14:15:42 CET] Isnt it better to actually output vfr then [14:16:59 CET] wm4, ffms2 has a lot of hacks for NVOPs like that [14:17:04 CET] the api is crap/broken for them [14:17:37 CET] nevcairiel: maybe, but that makes timestamp handling really fucking hard [14:18:06 CET] unless you pass the timestamp as DTS, which is wrong and won't work for some hwdec wrappers (although these wrappers might of course treat these differently again) [14:20:52 CET] now I'm wondering if it'd be possible to use PTS to pass through the timestamp [14:21:09 CET] somehow waiting until enough frames inherent to the codec delay are decoded, and then sort? [14:21:34 CET] but that doesn't handle dropped frames either [15:03:03 CET] ffmpeg 03Carl Eugen Hoyos 07release/2.8:aa3101a9e825: lavf/rtpenc_jpeg: Less strict check for standard Huffman tables. [15:04:10 CET] ffmpeg 03Carl Eugen Hoyos 07master:d208acf2c82b: lavf/rtpenc_jpeg: Less strict check for standard Huffman tables. [15:55:54 CET] fuck, wrong patchset [15:58:10 CET] wtf is my git send email doing [15:59:13 CET] dammit [16:00:00 CET] i don't get it [16:00:07 CET] i checked for the 3rd time [16:00:59 CET] oh well it seems it finally landed [16:01:16 CET] Action: Daemon404 always format-patch manualyl and send-emails that [16:02:03 CET] that's what i did [16:02:20 CET] dirty dir maybe [16:02:21 CET] i guess i messed up once, but why it sent 2 empty compsed messages is beyond me though [16:02:31 CET] beats me [16:29:23 CET] About the experimental flag, I'm not going to be available for interactive chat so I'll take this discussion to the ML [16:29:29 CET] (AAC experimental flag) [18:42:30 CET] turns out the Dirac decoder probably won't decode Low Delay profile properly or at all [18:42:42 CET] I think no one actually tested whether it works [18:43:19 CET] I'll make a patch to put av_log_ask_for_sample() if the file profile is Low Delay and put it up on the ML [18:44:34 CET] the Dirac specifications say one should read slices_x and slices_y only if the profile is Low Delay [18:45:24 CET] which the low_delay codepath doesn't [18:48:19 CET] ffmpeg 03Tobias Rapp 07master:bc517be6b0fb: Makefile: add cleanup of pkg-config files created by configure to distclean [18:51:52 CET] nevermind, low delay files should work perfectly [18:52:14 CET] because I don't believe any exist and only hawdware encoders could generate them [20:10:06 CET] [13FFV1] 15dericed opened pull request #16: Typos and cleanup (06master...06typos-and-cleanup) 02http://git.io/vRtUs [20:42:34 CET] ffmpeg 03Andreas Cadhalpun 07master:fb1bf4454e6e: git-howto: mention how to clone ffmpeg-web [21:10:15 CET] ffmpeg 03Ronald S. Bultje 07master:1ac89869db05: vp9: always keep s->bytesperpixel and ctx->pix_fmt in sync. [21:54:27 CET] ffmpeg 03Michael Niedermayer 07master:cd7524fdd13d: avcodec/apedec: Check length in long_filter_high_3800() [21:54:28 CET] ffmpeg 03Michael Niedermayer 07master:f03c2ceec174: avutil/mathematics: return INT64_MIN (=AV_NOPTS_VALUE) from av_rescale_rnd() for overflows [21:54:29 CET] ffmpeg 03Michael Niedermayer 07master:97b8db334a63: avutil/rational: Test av_rescale_rnd() with combinations of "special" values [22:45:24 CET] ffmpeg 03Alex Agranovsky 07master:79103f219903: avformat/mpjpeg: allow processing of MIME parts without Content-Length header [22:45:25 CET] ffmpeg 03Alex Agranovsky 07master:259c71c199e9: avformat/mpjpeg: utilize MIME boundary value to detect start of new frame [23:18:38 CET] trac is the new IRC [23:19:20 CET] we could have a separate channel for such notifications [23:20:02 CET] to not disturb the silence? [23:23:44 CET] sometimes it can go the other way, and such notifications will kill any real initiaior of possible chats [23:23:44 CET] doesn't bother me. i was just trollin. [23:24:12 CET] (because the signal noise ratio is too bad) [23:24:46 CET] on the other hand it sometimes initiates interesting discussions [23:25:01 CET] it raises topics (just like now :D) [23:25:03 CET] i like it [23:26:52 CET] i don't think i'd watch a fflog channel since i already have ffmpeg-cvslog and ffmpeg-trac ML [23:44:28 CET] ffmpeg 03Paul B Mahol 07master:b2517b02d990: avfilter/af_compand: do not clip; allow >0dB curve points [23:48:48 CET] I don't understand carl [23:49:00 CET] he backported the fix but keeps the bug open? [23:50:41 CET] really, the bot should post the comment text, and replying to the bot should post to trac [23:56:33 CET] is complex stuff from aemphasis ok to commit? [00:00:00 CET] --- Thu Dec 3 2015 From burek021 at gmail.com Fri Dec 4 02:05:02 2015 From: burek021 at gmail.com (burek) Date: Fri, 4 Dec 2015 02:05:02 +0100 (CET) Subject: [Ffmpeg-devel-irc] ffmpeg-devel.log.20151203 Message-ID: <20151204010502.B8E1B2AD6041@apolo.teamnet.rs> [00:21:48 CET] ffmpeg 03Michael Niedermayer 07master:d872643cfe07: avformat/utils: Check AVFormatContext->duration in compute_chapters_end() before use [00:21:49 CET] ffmpeg 03Michael Niedermayer 07master:ec7a3be11ed3: avformat/utils: Move end_time1 AV_NOPTS_VALUE Check after rescale [00:21:50 CET] ffmpeg 03Michael Niedermayer 07master:26379d4fddc1: avcodec/vp3: ensure header is parsed successfully before tables [00:32:29 CET] ffmpeg 03Ganesh Ajjanagadde 07master:1bb7db217d90: avutil/crc: avoid needless space wastage of hardcoded crc table [00:32:30 CET] ffmpeg 03Ganesh Ajjanagadde 07master:fa5d299496c1: avfilter/vsrc_mptestsrc: use lrint instead of floor hack [00:32:31 CET] ffmpeg 03Ganesh Ajjanagadde 07master:c6bea81acfa4: avfilter/vf_perspective: use lrint instead of floor hack [00:32:32 CET] ffmpeg 03Ganesh Ajjanagadde 07master:d64b6c38198e: avfilter/af_flanger: use rint instead of floor hack [00:53:33 CET] so I need to use RLE decoding but my escape level doesn't fit into an int8_t. how should I use the RL api? [00:54:01 CET] durandal_1707, perhaps? [00:54:04 CET] needed in cfhd [00:59:15 CET] wtf but RL_VLC_ELEM is an int16_T [01:11:13 CET] nevcairiel: when "configure: Replace `pr` since it is Nothing4You provided by busybox" is pushed to Libav don't merge it in FFmpeg; I'll send a patch for it after that is pushed [01:11:27 CET] oops not sure where "Nothing4You" came from [01:12:27 CET] But it's "configure: Replace `pr` since it is not provided by busybox" [01:16:53 CET] hmm seems dv does a trick to avoid that [01:17:00 CET] but with eugh global state [01:32:14 CET] Timothy_Gu : did you already change your "Nothing4You" password? :P [01:38:34 CET] it is more likely to be auto correction / replacement feature. [02:01:40 CET] it's a tab completition for a user here [02:06:49 CET] Nothing4You : identify yourself :P [02:07:01 CET] ahh hes in every channel [02:19:24 CET] llogan: yeah figured. there was a tab character after "not," and when copying irssi just autocompleted [02:44:11 CET] ffmpeg 03Michael Niedermayer 07master:0afdfbe11678: avcodec/jpeg2000: fix type of arguments of tag_tree_size [04:11:18 CET] ffmpeg 03Michael Niedermayer 07master:b46dcd5209a7: avutil/timecode: Fix fps check [05:40:59 CET] ffmpeg 03James Almer 07master:9ac5beaa86b5: avformat/mpjpegdec: fix mixed declarations and code [09:13:29 CET] meh how did I overlook that one [10:22:05 CET] ffmpeg 03Hendrik Leppkes 07master:46db068c5957: tests/api: Fix API test build on windows with --enable-shared [11:08:47 CET] ffmpeg 03Paul B Mahol 07master:acc2347cf47b: avfilter/af_agate: prepare for adding sidechain version [11:08:48 CET] ffmpeg 03Paul B Mahol 07master:bd5afecdcbb6: avfilter: add sidechaingate filter [11:08:50 CET] ffmpeg 03Paul B Mahol 07master:6907046130ee: avfilter/af_agate: add level_sc option for sidechaingate filter [11:08:50 CET] ffmpeg 03Paul B Mahol 07master:1b22bdf4e350: avfilter/af_agate: compile agate only when requested. [11:08:51 CET] ffmpeg 03Paul B Mahol 07master:4a43e559e1b3: avfilter/af_sidechaincompress: kill init function [11:08:52 CET] ffmpeg 03Paul B Mahol 07master:fff7f2df31e9: avfilter/af_agate: change default for detection to rms [12:04:03 CET] nice, vp9 hwdec patches [12:04:22 CET] any hardware which can use dxva VP9 hwaccel? [12:05:03 CET] intel braswell and skylake [12:06:08 CET] skylake, really? is it full hw decode? [12:06:19 CET] no hybrid only afaik [12:06:32 CET] oh ok [12:06:35 CET] nvidia seems to also be working on adding hybrid decoding, but their latest drivers still seem broken [12:07:25 CET] nevcairiel: hmm there was the product manager at vdd [12:07:25 CET] for nvidia [12:07:48 CET] maybe j-b could give you his contact details [12:07:51 CET] seemed like a nice guy [12:08:30 CET] i kinda expect a new nvidia driver this month, maybe they already finished it then [12:09:13 CET] the spec from MS is relatively new as it is, so maybe they worked with a pre-release spec and needed to update for the final [13:52:28 CET] michaelni: so why exactly did you try to make codec and hwaccel registration O(1)? this doesn't look very thread-safe [13:52:47 CET] even though the previous code attempted so by using avpriv_atomic_ptr_cas [13:57:56 CET] what is non thread safe on it ? [13:58:07 CET] the previous code was not O(1) [13:58:17 CET] it always iterated over the whole list [13:58:43 CET] I mean, it was somewhat thread safe [13:59:01 CET] I suppose the new code could also be considered thread safe if you assume that no atomics means weak atomics [13:59:35 CET] how much did this speed it up? [14:00:29 CET] it was meassureable IIRC, i dont remember the exact value, but with lets say 1000 codecs the old code would do i belive ~ 500 000 iterations [14:10:14 CET] and I suppose we remove this lock manager stuff only when all codecs have thread safe init [14:10:31 CET] instead of using a global mutex [14:15:36 CET] what codecs have nonthread safe init? [14:16:39 CET] all which don't have FF_CODEC_CAP_INIT_THREADSAFE set in AVCodec.caps_internal [14:16:48 CET] but most of them are maybe actually safe [14:21:07 CET] dv is nonthreadsafe [15:16:29 CET] complex is in c99 [15:16:43 CET] durandal_1707: you should specifically check for complex.h presence, I know of a bunch of systems that have cabs in math.h, but no complex.h [15:17:08 CET] huh? [15:18:06 CET] how It's supposed to work? [15:24:32 CET] they just have a small subset of complex functions in math.h [15:26:49 CET] what about tgmath.h? [15:28:06 CET] tgmath.h is generally not present on windows at all [15:28:46 CET] It should be, starting with VS2013 [15:28:51 CET] ? [15:29:18 CET] dont use tgmath please [15:29:20 CET] BtbN: no, the C tgmath.h header is not present even today [15:29:21 CET] it is an abomination [15:29:29 CET] it cant even be implemented in C [15:29:32 CET] (until C11) [15:30:11 CET] is available, but thats C++ [15:31:21 CET] C99 support in VS2015 is nearly complete, except tgmath.h and iirc some slightly different preprocessor behavior [15:31:50 CET] VLAs [15:32:35 CET] considering they are no longer mandatory in future C versions, thats a thing we can all do without =p [15:32:43 CET] theyre evil anyway [15:32:46 CET] so is tgmath [15:32:50 CET] nothing of value was lost [16:48:10 CET] hahahaha [16:48:12 CET] ahahahah [16:49:04 CET] so? [16:49:22 CET] Can he not read or do I misunderstand that commit message? [16:49:50 CET] J_Darnley, system libraries wont ship with libav abi compat enabled [16:50:17 CET] Oh [16:50:25 CET] I thought firefox hated ffmpeg. [16:50:50 CET] Why do they suddenly want to use a system lavc? [16:51:03 CET] they switched from gst to libavcodec i think recently? [16:51:06 CET] i thought someone said that [16:52:04 CET] I wonder if that means they're going to support h264 on XP now. [16:53:06 CET] who gives a crap? [16:53:51 CET] people who stick to an unsupported os can use unsupported software. [16:53:52 CET] there is more than just that int for abi compat [16:53:57 CET] yeah i know [16:54:24 CET] Well all browsers should have just used a system lavc from the begining and ignored the codec war shit. [16:59:47 CET] heh, so abi compat was useful in the end ;) [17:00:10 CET] i think that guy is just weird [17:00:25 CET] what does he want, ship a b inary that works on all possible systems without recompiling? [17:00:30 CET] thats unrealistic at best [17:00:40 CET] chrome ships their own libs on every os [17:00:43 CET] the only sane thing to do. [17:00:58 CET] lunuxers will hate you for it, but yes [17:01:29 CET] Excuse me, static linking into one giant exe is the sanest! [17:01:46 CET] thats not legal in this case [17:02:06 CET] the one case where i'd prefer a system library is crypto. [17:02:13 CET] s/^/also, / [17:02:20 CET] yeah leaves security concerns to the system and nto you [17:02:21 CET] :) [17:02:42 CET] Only cause google doesn't want to open their spyware for all the world to see. [17:04:27 CET] J_Darnley, well also because of flash [17:06:50 CET] Surely that sounds like a feature to people else days. [17:07:20 CET] "Flash deprecated" [17:07:38 CET] *sounds of joy heard all over the world* [17:08:44 CET] why is that even a problem for Firefox, do they really provide generic binary packages instead of shipping through distribution channels? [17:08:51 CET] open-source much? :) [17:14:27 CET] nevcairiel: yes I think there are binary Linux builds for FF [17:15:01 CET] that must be super painful code to try to support various ffmpeg/libav versions in a binary level [17:15:06 CET] at compile time is already terrible [17:15:10 CET] that's what they're trying [17:15:17 CET] they dlopen libavcodec [17:15:29 CET] well $work dlopen's avcodec [17:15:39 CET] why? [17:15:58 CET] avoids linking, its not used very often unless you watch a video [17:16:13 CET] but we still ship a specific version and compile against its headers [17:16:48 CET] even then you'd be better off dlopening your own wrapper [17:17:12 CET] why [17:17:15 CET] > There was never binary compatibility between FFmpeg and Libav, unless FFmpeg was compiled with a special configure flag not enabled by default. [17:17:22 CET] wm4: complete* binary compat [17:17:37 CET] the flag was just for stuff that would make it incompatible with ourselves [17:17:44 CET] limited scope [17:17:57 CET] rest was abi compat [17:18:58 CET] ok, changed it to avoid flames with certain assholes [17:20:43 CET] i find it hard to believe there are millions of people downloading ff binaries direct from mozilla on linux [17:20:55 CET] there probably aren't [17:21:05 CET] Action: J_Darnley must resist making a trac account and shit posting [17:54:22 CET] ffmpeg 03Timo Ter�s 07master:64f7db554ee8: mpegencts: Fix overflow in cbr mode period calculations [17:59:58 CET] what i find odd is that ffmpeg tells the distros how to distribute ffmpeg and they ignore us, but someone like mozilla will come along and get it fixed within a week [18:00:01 CET] i'm not bitter. :( [18:02:56 CET] michaelni : probably you should reply to that trac ticket, https://trac.ffmpeg.org/ticket/5057 [18:06:53 CET] Compn: where did FF get a fix? [18:10:59 CET] i'm saying they could get a fix from the distros, easily [18:11:09 CET] just another distro patch to revert that commit [18:11:33 CET] can't do that, the API would be different [18:11:36 CET] nevcairiel : why not offer some assistance to this bug reporter ? [18:12:05 CET] what kind of assistance [18:12:22 CET] i dont know, figure out what needs to be changed in firefox code maaaaybe [18:12:24 CET] hrmblegh [18:12:43 CET] so & [18:12:51 CET] Im not saying we should introduce libav compat again [18:12:52 CET] but [18:12:55 CET] then write this information down for other projects, due to api change. [18:13:09 CET] it may be worth hearing out this mozilla dude and see if he has any more practical thoughts or suggestions on this subject [18:13:21 CET] right [18:13:24 CET] mozilla is a pretty cool user to have, we already have chrome [18:13:33 CET] you have to realize how absolutely awesome that is [18:13:36 CET] wm4 : firefox wont be the only project to have this trouble with the api change. [18:13:50 CET] it's not even an api change, just abi [18:13:55 CET] .... whatever [18:14:28 CET] also it makes us look like jerks if our replies in our trac are "dont care, its already decided" [18:14:37 CET] the type changes, so it's an API change, but a minor one [18:14:54 CET] Compn: that's wrong, nobody will have a problem with it, except those who want to play ABI games [18:16:05 CET] Action: Compn stares into the abyss [18:18:47 CET] the abyss stares back at Compn [18:46:43 CET] BBB: a project that tries to keep binary compatibility with a wide range of versions will be blocking every change we want to make, we already have those people =p [18:48:54 CET] imho its a rather silly goal, at least chrome is sensible enough to just ship their version with the browser [18:49:53 CET] I dont even want to see the amount of nonsense such compatibility layer would collect [19:07:11 CET] any opposition to dropping the AAC encoder experimental flag? [19:07:45 CET] so far the only discussion was whether to have coders !twoloop usable only with experimental compliance [19:08:02 CET] well the biggest concern was that those are still broken [19:08:09 CET] so if you push that patch with it, its fine with me [19:10:58 CET] atomnuker, did you or claudio look at coverity issue 1338325 ? [19:11:49 CET] iam atm checking over the coverity issues and that is i think the only aac one [19:11:59 CET] aacenc that is [19:12:42 CET] could you give me a link? [19:13:44 CET] just type 1338325 in the CID search field at the top right, or you have no coverity account ? [19:15:02 CET] if you need an account say so and ill send you an invite [19:15:24 CET] yeah, I don't have an account [19:15:32 CET] so and invite would be nice [19:15:56 CET] BtbN: http://lists.freedesktop.org/archives/libva/2015-December/003748.html [19:15:59 CET] hevc 10 bit decoding [19:16:34 CET] and direct vp9 decoding support [19:17:12 CET] atomnuker, invite sent, 50% chance it will be in your spam folder [19:17:19 CET] fritsch: "direct"? [19:18:05 CET] as opposed to? :) [19:18:27 CET] http://lists.freedesktop.org/archives/libva/2015-December/003749.html [19:18:28 CET] yes [19:18:43 CET] nevcairiel: here is the P010 handling [19:18:47 CET] it converts to rgba32 [19:19:12 CET] if anyone else needs/wants a coverity account for ffmpeg, ping me [19:19:13 CET] someone outside of MS uses P010? [19:19:39 CET] Daemon404: its just nv12 in 10-bit, its not unsurprising that its being used for hwaccels [19:19:46 CET] i suppose [19:20:17 CET] i should clean out my patches for p010 support in ffmpeg [19:20:37 CET] nevcairiel: yeah you should :-) we have tested them in kodi and wait for you sending them upstream [19:20:40 CET] i only did a rather simple implementation, enough to suit my own needs [19:20:41 CET] http://lists.freedesktop.org/archives/libva/2015-December/date.html <- here VP9 bits [19:21:13 CET] it's broxton only it seems [19:21:21 CET] did anyone know "broxton"? <- kabilake? [19:21:28 CET] maybe atom [19:21:44 CET] yes, broxton is the next atom [19:21:49 CET] nice :-) [19:21:50 CET] http://lists.freedesktop.org/archives/libva/2015-December/003756.html [19:22:15 CET] guess they dont do hybrid decoding in libva eh [19:22:24 CET] because vp9 hybrid works through dxva2 on braswell [19:22:39 CET] ffmpeg 03Nicolas George 07master:3ab1e5a48c53: lavf: add FFERROR_REDO to let demuxers return no packet. [19:22:40 CET] vp9 hybrid is also there for braswell on linux [19:22:40 CET] ffmpeg 03Nicolas George 07master:0bac7a436be6: lavf/flvdec: use FFERROR_REDO instead of AVERROR(EAGAIN). [19:22:41 CET] ffmpeg 03Nicolas George 07master:cb14d30240b0: lavf/mpeg: use FFERROR_REDO instead of AVERROR(EAGAIN). [19:22:42 CET] ffmpeg 03Nicolas George 07master:eb2e4fb6745b: lavf/lxfdec: use FFERROR_REDO instead of AVERROR(EAGAIN). [19:22:43 CET] ffmpeg 03Nicolas George 07master:085ab74972bc: lavf/mpegts: use AVERROR_INVALIDDATA instead of AVERROR(EINTR). [19:22:49 CET] there is a intel-hybrid driver on 01 org [19:23:04 CET] i thought BtbN is working on it - therefore I pinged him [19:24:08 CET] that broxton should be available right now :-) [19:43:12 CET] "memset(&sce->lcoeffs[0], 0.0f, 3072*sizeof(sce->lcoeffs[0]));" [19:43:24 CET] how is that supposed to overflow? [19:43:49 CET] sorry, out of bounds access [19:44:33 CET] why memset 0.0f? [19:44:37 CET] why not just memset 0? [19:47:54 CET] it shouldn't matter, memset takes an int so the compiler should convert it to an int [19:48:11 CET] though I'll change it once I figure out how to applease coverity [20:08:43 CET] BBB: ping [20:09:00 CET] pong [20:09:11 CET] atomnuker: ty [20:09:38 CET] so, atomnuker and I are adding 10-bit decoding to dirac. What's the best way to have mixed 8/10-bit functions like in h264? [20:10:20 CET] what do you need dirac for? [20:10:29 CET] various things [20:10:41 CET] he's british [20:10:43 CET] it's in his blood [20:10:43 CET] I always thought it was a dead codec [20:11:16 CET] [13FFV1] 15michaelni pushed 5 new commits to 06master: 02http://git.io/vRZlT [20:11:17 CET] 13FFV1/06master 1482bbb6d 15Dave Rice: typos [20:11:17 CET] 13FFV1/06master 14bad4526 15Dave Rice: more readable descriptions of operators [20:11:17 CET] 13FFV1/06master 14723fbd0 15Dave Rice: capitalize Range coder when used [20:11:37 CET] huh? [20:11:42 CET] kierank: ehm... [20:11:54 CET] is that the ffv1 spec repo? [20:11:56 CET] odd [20:12:07 CET] kierank: uh & so & I think templating is easiest [20:12:21 CET] kierank: but it really depends on how much you care about performance vs. binary size [20:12:29 CET] templating of dsp functions I mean [20:12:37 CET] how does it work with structs [20:12:37 CET] https://github.com/FFmpeg/FFmpeg/blob/master/libavcodec/diracdsp.h [20:12:47 CET] and then the rest of the code would just take an extra bytes per sample param [20:12:50 CET] is there a way of reusing the function pointers? [20:12:52 CET] oh [20:13:08 CET] BBB yes its ffv1, i wonder why KGB doesnt show all 5 commits when there are 5 btw [20:13:24 CET] function pointers should be eay [20:13:31 CET] just use uint8_t everywhere [20:13:38 CET] and in the 10bpp functions internally, cast them to uint16_t [20:13:42 CET] ah you just change it [20:13:43 CET] ... [20:13:44 CET] stride would always be in bytes, not pixels [20:13:46 CET] so simple when you know how [20:13:56 CET] and then it just works [20:14:01 CET] theres a little casting in the dsp.c files [20:14:02 CET] but its ok [20:14:07 CET] we have helper templates for that [20:14:22 CET] bit_depth_template.c [20:14:27 CET] should be fairly helpful [20:14:34 CET] I believe vp9 and h264 both use it [20:14:48 CET] hevc also [20:14:58 CET] there's not much to change actually, it's mostly dsp functions that need changing in dirac [20:15:03 CET] right [20:15:22 CET] so its a little magic, but I think in the dsp, your function changes from uint8_t *px to uint8_t *_px [20:15:31 CET] and then you cast pixel *px = (pixel *) _px; [20:15:37 CET] where pixel comes from bit_depth_template.c [20:16:15 CET] I think it's only 8/10-bit that exists for now so it should be simple [20:16:25 CET] just the "next one up" for each integer type [20:19:06 CET] right, well, 8 to 10 is hardest [20:19:10 CET] because everything changes [20:19:13 CET] 10 to 12 is super trivial [20:22:35 CET] nevcairiel: I added chech for complex funcs similar to one for math funcs [20:26:08 CET] 10bit support for dirac decoder? [20:27:07 CET] durandal_1707: it's for the new hq profile [20:27:29 CET] Which is 10-bit and also slightly different [20:48:05 CET] ffmpeg 03Michael Niedermayer 07master:32bf6550cb9c: avformat/riffdec: Initialize bitrate [20:48:06 CET] ffmpeg 03Michael Niedermayer 07master:4e31176e145f: avformat/riffdec: remove special case for bitrate > 32bit [21:14:05 CET] nevcairiel: http://sprunge.us/GbiR which is faster the assignment or the memcpy? [21:15:48 CET] any half-decent compiler would hopefully replace the memcpy with a simple assignment in such a case [21:17:29 CET] hehe :-) [21:17:32 CET] exactly this happens [21:18:17 CET] nevcairiel: https://github.com/FernetMenta/xbmc/commit/adfa2b8c978754f3824b6a0cb46cd44ca9c43692#commitcomment-14772074 <- we had a major discussion here [21:18:30 CET] and our ARM devs jumped in [21:18:42 CET] and protested against the memcpy [21:20:13 CET] seems like mostly you discussing with yourself [21:20:50 CET] read on bottom of the page :-) [21:21:22 CET] sometimes it just avoids ugly casting or whatnot to use memcpy [21:21:45 CET] that was the original intention [21:23:39 CET] I'd rather question why there's truehd specific code in some big generic audio function (?) [21:24:00 CET] cause our syncer needs 24 packages [21:24:19 CET] thats just how truehd passthrough works, the frames are so small that the spec says to pack 24 together [21:24:46 CET] which gives a 20ms buffer, iirc [22:10:14 CET] michaelni: ping, have a question [22:17:35 CET] kierank, just ask, ill awnser when ive time / am done with what i work on atm [22:17:50 CET] it's a very historical question [22:18:12 CET] Any idea why this contradicts the spec [22:18:13 CET] https://github.com/FFmpeg/FFmpeg/blob/master/libavcodec/diracdec.c#L656 [22:18:20 CET] https://usercontent.irccloud-cdn.com/file/66bkVKzQ/ [22:18:22 CET] spec ^ [22:18:26 CET] https://lists.ffmpeg.org/pipermail/ffmpeg-soc/2008-December/006067.html [22:18:30 CET] commit that changed it [22:24:26 CET] why does it contradict the spec ? isnt that commit just a simplification [22:25:11 CET] spec says level 0 only has LL [22:25:22 CET] whereas code appears to do 0,1,2,3 for level 0 [22:25:33 CET] for orientation [22:25:50 CET] maybe the spec just 1-indexes everything? [22:25:56 CET] that would explain the code we have [22:26:13 CET] it doesn't [22:26:17 CET] level 0 gets 0-3, level 1+ just get 1-3 [22:26:47 CET] there are other loops in the spec that loop through 0 [22:27:42 CET] the commit changes it from the spec version to another [22:27:45 CET] i suppose it seems wrong then, yes [22:27:47 CET] not that the spec for loop is inclusive [22:27:56 CET] note* [22:29:51 CET] but it seems like something that would result in rather noticeable problems [22:31:01 CET] the code does before and afterwards 0,1,2,3,1,2,3, ... i dont see the difference [22:31:33 CET] not for level=0 [22:33:23 CET] according to the spec its [0] = {0}, [1] = {1,2,3}, etc, the old code did this, the new code does not, it reads [0] = {0,1,2,3}, [1] = {1,2,3} ... unless some other code was changed to treat level 0 differently and expect it to be read like this [22:35:14 CET] nevcairiel, actually there is hybrid decoding for libva but it's a separate driver (not the i965 one) [22:36:28 CET] kierank: the code handling these levels seems somewhat complex, he may just as well have decided to just put level 0 and 1 into the same struct, because they dont overlap [22:36:46 CET] it's a possibility yes [22:37:57 CET] hmm could be [22:37:58 CET] https://github.com/FFmpeg/FFmpeg/blob/master/libavcodec/diracdec.c#L833 [22:43:22 CET] TD-Linux: but the default driver was extended to load the hybrid .so files [22:43:49 CET] so for the future it can use all of those "third party hybrid thingies" via the same api [22:44:36 CET] + int cb_numx = s->codeblocksh[b->level + (b->orientation != subband_ll)]; [22:44:36 CET] + int cb_numy = s->codeblocksv[b->level + (b->orientation != subband_ll)]; [22:44:39 CET] seems offsetted here [22:49:10 CET] fritsch, ah cool. I built the hybrid driver but didn't get around to actually testing it yet. motivation is really low because ffvp9 is fast enough :) [22:55:44 CET] TD-Linux: see the latest libva ML messages [22:56:06 CET] TD-Linux: vp9 will be supported natively with next gen - so good chance to get it going in ffmpeg [22:56:08 CET] I didnt actually get a chance to properly benchmark the vp9 hybrid decoder on braswell yet [22:56:24 CET] i benchmarked hevc-10bit decoder for broadwell ... [22:56:26 CET] the ffmpeg_dxva2.c module is bad for performance [22:56:27 CET] nightmare :-) [22:56:52 CET] once its pushed and everything I should benchmark using my own code [22:56:52 CET] 4k 10 bit hevc you can just forget [22:57:08 CET] on those hybrid [23:00:12 CET] nevcairiel: ah just seen your vp9 hwaccel preperation [23:10:59 CET] preperation? Thats everything thats needed for dxva2 vp9 :) [23:15:21 CET] durandal_1707: that random number at the beginning of the table is the length [23:32:43 CET] hm, git send-email gets stuck when sending to ffmpeg-devel at ffmpeg.org [23:32:57 CET] or maybe it's my internet [23:34:16 CET] the ML wont be involved in any case, either your internet or your mail server [23:35:50 CET] just took a long time for some reason [23:48:36 CET] atomnuker: lcoeffs is only 1024 elements [23:49:29 CET] maybe you wanted ltp_state, wh ich is 3072 elements? [23:52:48 CET] left a comment on the ML, time for sleepz [00:00:00 CET] --- Fri Dec 4 2015 From burek021 at gmail.com Fri Dec 4 02:05:01 2015 From: burek021 at gmail.com (burek) Date: Fri, 4 Dec 2015 02:05:01 +0100 (CET) Subject: [Ffmpeg-devel-irc] ffmpeg.log.20151203 Message-ID: <20151204010501.8D4E02AD6040@apolo.teamnet.rs> [00:07:53 CET] Hi! I know this is a hot issue, but I have some code that uses libswresample - which is not available for Ubuntu 14.04. It works on my platform, but someone that helps me can't compile the code (obviously, due to the lack of libswresample). He says that there's libavresample-dev available. What should I do? Switch to libavresample? Or is there a more � standard ��way? [00:14:48 CET] ... There don't seem to be any difference between libavresample and libswresample, besides the function prefixes? O_o [00:16:38 CET] there is, lavr clips floats [00:17:09 CET] (which means?) [00:18:08 CET] you do not know? [00:18:41 CET] no? It distords sound? [00:18:57 CET] yes [00:19:20 CET] is it really that bad? Is libavresample still developed? [00:20:08 CET] I'm kind of lost, why are libsw/libav coexisting? Is it a side effect of libav/ffmpeg ��war ��? [00:24:14 CET] yes [00:49:13 CET] can someone help me with vdpau ( nvidia ) decoding [00:49:16 CET] how do i set that up [00:50:40 CET] i have nothing in : Supported hwaccels: [00:50:43 CET] Hello. [00:52:00 CET] Would the following be a reasonable command for converting things to prores with current ffmpeg builds? [00:52:01 CET] ffmpeg -y -probesize 5000000 -i inputfile.mov -c:v prores_ks -profile:v 3 -qscale:v 11 -vendor ap10 -pix_fmt yuv422p10le -s 1920x1080 output.mov [00:52:09 CET] I mean, it seems to work. [00:57:01 CET] prelude2004c: recompile with --enable-vdpau [00:57:50 CET] and/or install libvdpau-dev [01:01:17 CET] Hi there! I was following MS tutorial on how to compile FFMPEG for WinRT https://trac.ffmpeg.org/wiki/CompilationGuide/WinRT and used these commands http://pastebin.com/W69G8aZv. However, I got cl is unable to create an executable file. C compiler test failed. [01:01:51 CET] Any ideas what could be wrong? [01:01:53 CET] Mantas_: hallo [01:12:15 CET] how does -to work with -ss ? because -ss start -to end doesn't trim the video between them [01:12:26 CET] i hate using -t because i have to calculate timestamp difference [01:13:46 CET] Hfuy: looks ok. not sure if qscale will do anything. might want to look into "-movflags +write_colr" [01:17:08 CET] dongs: are you using -ss as an input option? [01:17:32 CET] yeah [01:17:40 CET] i mean, -ss foo -t bar works fine [01:17:51 CET] but i have to calculate end-start and input it as t. [01:18:02 CET] i want to just do -ss start -to end [01:24:38 CET] dongs: https://trac.ffmpeg.org/wiki/Seeking#Cuttingsmallsections [01:25:02 CET] Action: Hfuy listens to Faithless's "Insomnia" [01:25:11 CET] llogan: yes, that does not work [01:25:25 CET] oh i guess i need -copyts [01:25:58 CET] no the first example from tehre definitely does NOT work. [01:26:06 CET] it cuts from 1:00 to end of clip [01:28:03 CET] will try with -copyts and see [01:29:48 CET] ffmpeg -ss 10:00 -i "2015.12.02 LIVE.ts" -to 12:00 -c copy ff.ts [01:29:51 CET] made 12 minute long ff.ts [01:29:53 CET] .... [01:30:05 CET] -copyts made 0-byte file cuz it sait it was "empty" [01:31:28 CET] i see, -ss after -i makes it work as intended [01:31:49 CET] but 3rd case with -copyts doesnt work, oh well, at least the 2nd one does. [01:34:23 CET] Seems to work for me [01:34:35 CET] maybe depends on container type [01:34:53 CET] Output file is empty, nothing was encoded (check -ss / -t / -frames parameters i [01:34:55 CET] f used) [01:35:05 CET] ^ thats what it does w/same line as above and -copyts aded [01:35:06 CET] ffmpeg -y -ss '00:05:52.5' -i viedo.mkv -to 6:22.6 -c copy -copyts ~/out.mkv [01:35:58 CET] The listed duration in the container is off though [01:36:02 CET] For whatever reason [01:39:35 CET] How would I go about finding whether the version of ffmpeg I have is 32 or 64 bit? [01:40:51 CET] ^ dongs [01:40:59 CET] Hfuy: file `which ffmpeg` [01:41:54 CET] cbsrobot-: yes beacuse one line of output really needs to be on pastebin. [01:42:16 CET] dongs: ... COMPLETE console output [01:42:23 CET] c_14: not on windows, I suspect [01:42:33 CET] ah, eh [01:42:41 CET] try right clicking and looking in properties? [01:43:23 CET] Not so far as I can see. [01:43:33 CET] cbsrobot-: http://bcas.tv/paste/results/2MxbNq14.html shruggin' [01:44:23 CET] dongs: Duration: 00:49:03.52, start: 3692.436200, bitrate: 9781 kb/s [01:44:28 CET] Hfuy: copy it to a unix machine and run file on it... [01:44:29 CET] yeah? [01:44:38 CET] your start ts is not 0 [01:45:02 CET] it mostlikely never will be, with mpegts [01:45:03 CET] and? [01:45:11 CET] I guess that's why your output is wrong [01:45:12 CET] dongs: try adding -start_at_zero [01:45:29 CET] Ah. Yes. I have a bit-depth-incompatibility problem. [01:45:32 CET] Action: Hfuy scratches his head [01:45:46 CET] c_14: nope, same result as w/o, 12minute long cut [01:46:06 CET] 64-bit ffmpeg can't talk to 32-bit avisynth [01:46:06 CET] dongs: with -copyts [01:46:22 CET] that worekd. [01:46:32 CET] \o/ [01:46:52 CET] thanks. [01:47:43 CET] hey can anyone help me with how to get -hwaccel vdpau for fmpeg [01:48:05 CET] compile with --enable-vdpau [01:48:07 CET] i instaled the rpms and they are there. i compiled ffmpeg again and still shows no hardware accells available [01:48:11 CET] yup did that [01:49:29 CET] ./configure --enable-gpl --enable-nonfree --enable-nvenc --enable-pthreads --enable-libx264 --enable-debug --enable-static --enable-vdpau --enable-hwaccels [01:49:33 CET] that is what i have been using [01:55:04 CET] Gaak. How do I make it do 16-bit PCM stereo sound? [01:55:14 CET] -c:a pcm_s16le [01:55:25 CET] Thanks. [01:55:39 CET] prelude2004c: seems to work for me (didn't try your exact configure line because I don't have the nvenc headers) [01:56:01 CET] In configure's output, what's listed under "Enabled hwaccels:"? [02:11:30 CET] Action: Hfuy happily encodes prores quicktimes directly out of windows software [02:11:36 CET] Action: Hfuy gives apple the virtual finger [02:19:35 CET] What does "qscale" mean? [02:19:39 CET] Something to do with the quantiser? [03:52:05 CET] does -preset parameter work? I use FAST and SLOW and the slow file was larger. Same crf=20 same output format. [03:52:16 CET] s/use/used* [04:37:14 CET] also are crf leves different in libx265 vs libx264? is crf 23 on x265 the same quality as crf 23 on x264? [04:37:32 CET] yes, eh no, eh they're different [04:37:38 CET] i know both have different compression techniques [04:38:09 CET] c_14: is there a site that shows a table of the levels side by side? [04:38:26 CET] none I know of [04:42:06 CET] c_14: hevc looks like bad news the more I read about it. the fees are just out of control, aka a loose canon [04:42:35 CET] no wonder Open Media Alliance wants their own codec(royalty free) [04:42:44 CET] cisco has thor or daala. [04:43:48 CET] what's wrong with vp9 [04:43:54 CET] other than the speed [04:43:56 CET] not sure [04:44:09 CET] google is likely to pitch its vp9/vp10 ideas to the alliance [04:44:33 CET] should they not make vp9 usable before working on vp10 [04:44:50 CET] right now vp9 takes forever to encode anything... not sure why parallel is offline [04:45:07 CET] furq: every year they are pushing updates to it. [04:47:53 CET] does anyone know what bitrate netflix/amazon pushes their UHD content to viewers? [04:50:44 CET] furq: how do you find googles vp9? [05:37:54 CET] furq, well if you look at libvpx git most of the work is on vp9 [05:39:07 CET] a lot of people are happy with vp9 now. but many others want a) input into the design other than 100% google b) it standardized c) better guarantees that it's royalty free [05:40:00 CET] also vpxenc uses multiple cores by default now [07:23:13 CET] Hi guys [07:24:28 CET] I have a gif I want to render in opengl, so what I did was using imagemagick generate new images, load them as textures, and render. [07:24:40 CET] It works fine, but messes up the FPS [07:25:23 CET] another way I could retain the fps is by converting the gif into an mp4, and decomposing *that* mp4 into images and loading them [07:26:03 CET] granted this will generate too many duplicate frames, but let's say I am not worried about that for now. [07:27:18 CET] So, how do I make am mp4 from a gif, with an explicitly defined FPS? [07:27:23 CET] *an [07:28:53 CET] the same way you would make an mp4 from anything else [07:31:26 CET] Action: nilspin is going through the docs [07:33:08 CET] there's no reason to go via mp4 if you just want images at a certain framerate [07:33:31 CET] ffmpeg -i src.gif -r 60 dest%04d.png [09:11:23 CET] furq, thank you. However, I didn't understand the 'dest%40d.png' part (linux noob here :) [09:13:25 CET] nilspin: hallo noob [09:13:38 CET] waressearcher2, hi. [09:15:45 CET] ah, its just a format specifier. Stupid me indeed. [11:06:37 CET] hi [11:07:59 CET] i am using aconv, converting from a webm file (~20 minutes long) to mp4 (usually less than 1 minute long). as i play back the mp4 in vlc, the progress bar doesn't progress normally. for example it'll stay on 0, and then only jump to 35 seconds, and then to 48 seconds, and then the video has ended. how do i make this work better? [11:12:41 CET] I think you have an old version of it, although I will have to mention that avconv comes from Libav, not FFmpeg :) [11:12:49 CET] #libav would be the correct place for questions regarding it [11:17:50 CET] i thought that was the same thing [11:18:10 CET] well in the end maybe someone knows anyways [11:18:39 CET] my version is from march, so not that old [11:18:46 CET] it's the latest my distro has [11:25:13 CET] build date doesn't matter [11:33:55 CET] oh right right [11:33:59 CET] ummmmm version 9.18-6:9.18-0ubuntu0.14.04.1, [11:36:09 CET] not here, on the other channel you're on, which is the right forum for support regarding that thing. or you can grab a static FFmpeg ffmpeg build from one of the linked downloads [11:54:39 CET] Hello, how can I check an MP3 file for defects, and is there a tool or way to fix any defects without re-encoding the whole file? I have an MP3 which Clementine reports as being 13 hours long @ 40kbps, while ffplay shows it as 4 hours long @ 128kbps. [11:54:48 CET] ffplay's assessment sounds more correct [11:56:09 CET] the file in question: http://slayrated.org/stream/raw-2015-12-01.mp3 [12:13:22 CET] DrSlony: does the file have a xing header? [12:14:05 CET] DrSlony: and mp3diags should be able to add/fix the vbr header [12:21:21 CET] JEEB: thanks [12:51:01 CET] when calling avformat_write_header I get "mpegts dimensions not set", I have set w/h in codec and would assume avformat_new_stream() would fill it, what am I missing? [12:52:14 CET] jasom doesnt look like it has a xing or vbri header [12:54:41 CET] thank you for mp3diags [15:16:57 CET] why does CONFIG_FFPLAY comment out in make file when disable-everything but dont disable programs [15:17:02 CET] Arch=arm [15:17:09 CET] where as arch=x86 it does not happen [15:17:34 CET] why does it disable even when i go --enable-ffplay [15:18:41 CET] shared libraries [15:19:08 CET] disable debug and doing release [15:20:50 CET] ffplay requires enabling output devices like SDL so it can play back the video [16:03:46 CET] hey anyone around... i am having trouble getting vdpau installed with ffmpeg.. i installed the rpms ( libvdpau-devel-0.4.1-1.el5.rf.x86_64 & libvdpau-0.4.1-1.el5.rf.x86_64 ) and i also did --enable-vdpau on compile of ffmpeg.. using version N-76957-g7b11eea .. not sure what else to do.. nothing shows up for -hwaccels and i need to decode using the GPU .. any help would be great [16:06:45 CET] After the ./configure, what is listed under "Enabled hwaccels:"? [16:38:02 CET] Hi. Is it normal that ffprobe outputs "display_picture_number=0" for every frame? without it there is no way to tell which frame it is actually analyzing, or is "coded_picture_number" always the same as the "display_picture_number" in H264 streams? [16:53:27 CET] I believe it means the container doesn't offer a frame counter feature or that it can't be reliably measured without explicitly counting them [17:01:21 CET] DHE thanks for the information. by "explicitly counting" you mean decoding, which would take forever? because it already goes through all frames with ffprobe and does not "count" them [17:02:38 CET] but you couldn't seek to a random frame and say "this is frame #5123" with certainty [17:03:43 CET] hey c14 sorry had a meeting [17:03:48 CET] nothing shows under enalbed hwaccels [17:03:50 CET] thats the problem [17:04:31 CET] DHE certainly not, but i had it start from the beginning and analyze the whole thing [17:06:52 CET] Admin_234234: check config.log [17:27:32 CET] hey guys. [17:27:33 CET] hey anyone around... i am having trouble getting vdpau installed with ffmpeg.. i installed the rpms ( libvdpau-devel-0.4.1-1.el5.rf.x86_64 & libvdpau-0.4.1-1.el5.rf.x86_64 ) and i also did --enable-vdpau on compile of ffmpeg.. using version N-76957-g7b11eea .. not sure what else to do.. nothing shows up for -hwaccels and i need to decode using the GPU .. any help would be great [17:27:58 CET] c14.. config.log is stored where ? [17:28:05 CET] in the same compile directory.. checking [17:44:33 CET] hey C14... /usr/include/vdpau/vdpau_x11.h:44:22: fatal error: X11/Xlib.h: No such file or directory [17:44:42 CET] can i get that library installed without actually installed the UI ? [17:45:45 CET] Realistically, are there any properly made live streaming applications? [17:46:02 CET] Most of what I've seen look like command line shoddy mickeymouse implementations. [17:46:24 CET] what have you seen? [17:46:37 CET] i think i just answered my own question :) [17:49:24 CET] klaxa|work, gstreamer /ffmpeg plugins/daemons to act as an http server [17:55:07 CET] for what, HLS/dash? [17:55:37 CET] in general, both of those are designed to be served directly as files from a standard http server, no special server logic needed. [17:56:14 CET] so the hls or dash streaming stuff does basically consist of dropping files into a directory to be picked up by an http server, possibly with some cdn in front of it. [17:58:44 CET] so another issue... now i got it compiled [17:58:45 CET] Error while decoding stream #0:7: Invalid data found when processing input [17:58:45 CET] Cannot open the X11 display . [17:58:45 CET] VDPAU init failed for stream #0:7. [17:59:26 CET] some of the gstreamer folks were working on an all-in-one streaming server "gst-streaming-server" which did hls/dash/rtsp, maybe other formats, but i don't think it really went anywhere. Not sure of the current state. [18:15:30 CET] can anyone help ? i am running a linux shell .. i compiled it ok with having the x11_devel installed but it still will not connect to use the card [18:15:38 CET] does this mean i have to actually install the UI for the system ? [18:15:48 CET] i dind't want to do that as i don't need it at all and use everything in ssh [18:15:52 CET] can anyone assist ? [18:19:45 CET] as far as I know, vdpau is only available when running inside X, which is kind of a pity. what is your intended use? nvidia gpu encoding? [18:21:06 CET] if so, you'd probably want to look at using the nvenc rather than vdpau [18:24:42 CET] gpu decoding [18:24:50 CET] i already use nvenc and it is working well [18:24:56 CET] the problem i am having is on the decoding side [18:25:08 CET] i dont' have enough cpu power to decode h264 correctly [18:25:14 CET] so i need to offload the decoding to the GPU too [18:25:23 CET] the M4000 has both a decoder and encoder built in [18:37:30 CET] help.. i need somebody .. help.. not just anybody.. help.. i need someone .. HELLLLPP !! [18:37:31 CET] :) [18:58:27 CET] I now have my code to encode with h264 then mux to mpegts, the problem is that I'm probably doing something wrong with avframe->pts, when I was creating a raw h264 I set QFrame->startTime() and everything went well, but mpegts it seems to want an increasing number which creates a "1sec" video [18:58:41 CET] how should I deal with pts? [18:59:18 CET] I was setting pts to 1 on first frame, 2 and so on... [18:59:30 CET] anyone know centos ? vdpauinfo: cannot connect to X server :0.0 .. i am on the home strech and can't get it going [18:59:36 CET] it doesn't show any warnings but the output is odd [18:59:41 CET] centos channel is dead [19:05:31 CET] prelude2004c: you're logged into the right server as the right user for an existing X11 session? [19:08:23 CET] i am root [19:09:14 CET] i installex x11-devel and thats aobut it [19:12:44 CET] dantti, MPEG-TS should probably have 1/90000 timebase :) [19:12:58 CET] meaning that frame at 0sec = 0pts, 1sec = 90000pts, etc. [19:13:06 CET] and time_base on stream should be set to 1/90000 [19:13:19 CET] Mavrik: 90000 instead of stream->time_base = (AVRational){ 1, 25 }; ? [19:13:30 CET] *instead of 25 [19:14:38 CET] and should I calculate the seconds my self? by probably using QFrame->startTime()? [19:16:31 CET] I have no idea what startTime() does ;) [19:16:35 CET] on the mux example there is ost->frame->pts = ost->next_pts++; which I did the same, so I'm unsure how to relate that :P [19:16:41 CET] You should set PTS so the timestamps on the frames will be good. [19:16:53 CET] PTS sets time at which the frame has to be displayed [19:17:00 CET] It's expressed in timebase units [19:17:07 CET] Returns the presentation time (in microseconds) when a frame should stop being displayed. [19:17:17 CET] Start or stop? :) [19:17:18 CET] *should start [19:17:29 CET] Well, then you should use that and convert it to your timebase [19:17:30 CET] copied the docs from the endTime :P [19:18:04 CET] (startTime / 1000 0000) * (90000 / 1) :) [19:18:21 CET] (Watch rounding and stuff) [19:19:06 CET] and (AVRational){ 1, 90000 }; right? [19:19:50 CET] Yep, that's the standard for the container, should be set on AVFormatContext too I think (keep mixing those up -_-) [19:23:59 CET] ok, that didn't work, maybe rounding is to blame :P [19:27:56 CET] dantti, for 25fps stream, the PTS has to be 0, 3600, 7200, etc. [19:29:29 CET] shouldn't it be 0, 25, 50? [19:34:06 CET] On what basis? :) [19:34:28 CET] 1, 15? [19:34:32 CET] 1, 25 [19:36:11 CET] I'm not sure how many frames will I get, so if I get 15 frames in a second pts is 1...15 then 26...50 ? [19:36:48 CET] dantti, but didn't you say you get frame timestamps from the source_ [19:36:50 CET] ? [19:36:55 CET] yup [19:37:39 CET] should I store that value from the first frame and then remove it as a base? [19:38:57 CET] hey can anyone help me.. i am sooo close but i can't figure out this last thing... using vdpau.. i have it all there but its saying tha tit can't connect to X server ; tried xhost + ; tried sshd_config x11forward = yes .. i am out of ideas [19:39:08 CET] dantti, you don't have to, you have to convert to right time units. [19:39:24 CET] prelude2004c, you need an active X server with a GPU driver enabled. [19:39:50 CET] how does one do that though .. i have never done this before [19:40:00 CET] this seems to be rounding to 0 always (qFrame->startTime() / 10000000) * (90000 / 1) [19:40:18 CET] yum install xorg-x11-server-Xvfb < ran this [19:40:24 CET] running on port 6001 it seems [19:40:31 CET] i have an nvidia M4000 card installed [19:45:51 CET] ok, startTime() is always -1, I guess that's due the source being the camera, Qt probably sets that on video playback [19:46:22 CET] I wonder why the raw h264 had it fine... [19:54:41 CET] Mavrik: I guess I need a reference clock then no, and check with a reference to know when a second has passed? for a variable fps should I just drop more than 25 frames or is there some better approach, since this will go through network I guess 25fps is a good number... [19:54:58 CET] Do you have variable FPS or static FPS? [19:55:19 CET] I think it's variable as I have no control over it [19:55:35 CET] in fact I think those are the frames displayed by the GPU [19:57:40 CET] Well do check [19:58:48 CET] not sure how [20:08:57 CET] ok, as it seems it's indeed variable [20:09:58 CET] but I can set a maximum frame rate it will have [20:11:15 CET] can someoen help me please. Cannot open the X11 display . [20:11:15 CET] VDPAU init failed for stream #0:5. [20:11:23 CET] Mavrik: for a variable fps I guess I need to calculate the frame time myself right? otherwise would just be jumping [20:11:34 CET] when i try to use it as -hwaccel .. using root on a centos 7 install .. no GUI [20:12:17 CET] I mean a static FPS is just multiples of itself I think [21:42:02 CET] that --enable-outdev=sdl made it work [21:42:11 CET] thanks whoever said that like 6 hours ago [21:44:02 CET] So since the [21:44:03 CET] Past duration 0.999992 too large [21:44:03 CET] Last message repeated 20 times [21:44:12 CET] message doesn't seem to actually do anything [21:44:15 CET] Can I disable it? [21:54:40 CET] anyone good with xrdp and can make it reuse session? [21:55:07 CET] everytime i switch port=-1 to port=3399 or any port not used it goes from working to not working [21:55:14 CET] all of firewall is destroyed and off [23:28:29 CET] Hello. I am trying to download a DRM free silverlight stream. It has a ism/manifest, but the url has a token. Whenever i put the entire url into ffmpeg with the token i get "invalid data found when processing input" [00:00:00 CET] --- Fri Dec 4 2015 From burek021 at gmail.com Sat Dec 5 02:05:01 2015 From: burek021 at gmail.com (burek) Date: Sat, 5 Dec 2015 02:05:01 +0100 (CET) Subject: [Ffmpeg-devel-irc] ffmpeg.log.20151204 Message-ID: <20151205010501.920942AD6004@apolo.teamnet.rs> [01:46:31 CET] I am trying to compile ffmpeg with x265 support and I am getting the following error in my config.log file: include/x265.h:103: error: expected specifier-qualifier-list before 'bool' [02:06:42 CET] I found my answer. [02:06:45 CET] Hi [02:08:56 CET] I got ffmpeg compiled successfully on Msys 64bit and x264 compiled, when I configure ffmpeg with libx264, I used extra-cflags and extra-ldflags to point to libx264 location but It always show the error "libx264 not found" [02:09:01 CET] hello [02:09:04 CET] anybody alive? [02:09:16 CET] Barely [02:09:35 CET] chaupv: Can you give your ./configure line? [02:09:41 CET] Ok [02:09:53 CET] thank wannabegeekstar [02:09:54 CET] wannabegeekster: do you know anything about how ffmpeg converts .txt files to video? [02:10:05 CET] ./configure --toolchain=msvc --arch=x86 --enable-yasm --enable-asm --enable-shared --enable-libx264 --enable-gpl --enable-static --extra-cflags="-O3 -std=c99 -fomit-frame-pointer -fno-strict-aliasing" --extra-ldflags="-Wl,-z,now -Wl,-z,relro,-z,noexecstack" --extra-cflags=-I/usr/include/ --extra-ldflags=-L/usr/lib/ [02:10:07 CET] im talking about txt files that are embedded in HLS playlists and contain plaintext [02:10:15 CET] wannabegeekster: you found that answer, but what was your answer? [02:10:41 CET] allfro: ffmpeg can "play" a txt file like any other input it supports [02:10:43 CET] aalfro: Are you wanting to convert text files or images to video? I have not seen anything on converting .txt files to video [02:10:57 CET] llogan: https://bitbucket.org/multicoreware/x265/issues/212/new-libx265-not-found-in-ffmpeg-configure [02:11:05 CET] Yes, i have never been converted text to video [02:11:13 CET] it sounds strange.. [02:11:15 CET] llogan: so ffmpeg can convert regular text to video frames? [02:11:16 CET] but how [02:11:23 CET] is it through the draw_font filter? [02:11:43 CET] chaupv: Are those libraries installed in /usr/lib and /usr/include? [02:12:03 CET] wannabegeekster, yes [02:12:35 CET] libx264 has already compiled and install in to /usr , as it's a prefix, --prefix=/usr/ [02:12:35 CET] chaupv: Are you on windows or linux? [02:12:49 CET] window, Msys64 [02:13:42 CET] allfro: via the ASCII/ANSI art decoder. libavcodec/ansi.c [02:13:50 CET] humm. I am not so familiar on Windows there chaupv. [02:14:11 CET] allfro: and tty demuxer [02:14:50 CET] wannabegeekster, thanks... [02:15:17 CET] if you have any tips, please give me.. [02:15:53 CET] PKG_CONFIG_PATH="$HOME/ffmpeg_build/lib/pkgconfig" ./configure --prefix="$HOME/ffmpeg_build" --extra-cflags="-I$HOME/ffmpeg_build/include" --extra-ldflags="-L$HOME/ffmpeg_build/lib" --bindir="$HOME/bin" --pkg-config-flags="--static" --enable-gpl --enable-nonfree --enable-libfdk-aac --enable-libfreetype --enable-libmp3lame --enable-libopus --enable-libvorbis --enable-libvpx --enable-libx264 --enable-libx265 [02:16:16 CET] That is what I am doing right now, but it is obviously quite a bit different than you right now. [02:17:08 CET] llogan: so for this http://pastebin.com/NacCAHPz [02:17:10 CET] wannabegeekster, so ffmpeg in your workspace can find x264, x264 well, right? [02:17:19 CET] Yes, that is correct. [02:17:29 CET] would that be the ansi decoder? [02:17:52 CET] wannabegeekster, but as you said, you are working on linux, not in window [02:17:58 CET] (hls playlist in bottom) [02:18:17 CET] Yes, so that makes it harder for me to figure it out. Although I am running windows and would be happy to try. Are you following a guide? [02:18:29 CET] I have the gnutools installed on my machine now already [02:19:00 CET] Yes, I follow many guidelines [02:19:37 CET] but there is no guide answer for the "libx264 not found" still now [02:20:42 CET] chaupv: look at your config.log file. That will give you much more information [02:21:07 CET] I looked at it many times already [02:21:17 CET] ... [02:21:25 CET] Can you pastebin.com it for me to look at? [02:22:18 CET] Ok, here is the last piece to show libx264 not found [02:22:19 CET] LINK : warning LNK4044: unrecognized option '/Wl,-z,relro,-z,noexecstack'; ignored LINK : warning LNK4044: unrecognized option '/LC:/environtment/msys64/usr/lib/'; ignored LINK : fatal error LNK1181: cannot open input file 'x264.lib' ERROR: libx264 not found [02:23:13 CET] Did you fix the issue with the /Wl? [02:24:08 CET] plus do a find / -name 'x264.lib' and make sure it is in the the right location. I assume you have done that already [02:25:37 CET] Wannagegeenkster, [02:25:41 CET] find / -name 'x264.lib' /usr/lib/x264.lib find: failed to read file names from file system at or below /: No such file or directory [02:27:29 CET] after I user you command to find x264.lib, found result is /usr/lib/x264.lib, then it also print out one more line: find: failed to read file names from file system at or below /: No such file or directory [02:27:40 CET] sorry, after I use.. [02:28:25 CET] finally [02:28:28 CET] i figured it out [02:28:34 CET] thanks llogan for your input [02:28:38 CET] damn he left [02:28:42 CET] or she [02:28:50 CET] thanks #ffmpeg [02:32:13 CET] This has a misspelling: /LC:/environtment/msys64/usr/lib/ [02:32:23 CET] environtment should be environment [02:32:36 CET] Not sure if you copy pasted that or typed it in [02:34:08 CET] yes, it's correct path, [02:34:40 CET] it's my purpose to make a little different name... [02:35:39 CET] ahh ok cool [02:35:58 CET] What about this specifically? warning LNK4044: unrecognized option '/Wl [02:36:48 CET] i've not used cl much but "unrecognized option '/LC:/environtment/msys64/usr/lib/';" suggests it's not being passed your library path [02:37:51 CET] furq, thanks.. [02:38:03 CET] apparently that's not even an option for msvc [02:40:03 CET] furq, what should i change? [02:42:24 CET] but furq, ffmpeg in window(msys) require libx264.lib, so I must to use cl and link to compile it [02:43:00 CET] https://trac.ffmpeg.org/wiki/CompilationGuide/MSVC#CompilingFFmpegwithexternallibraries [02:43:03 CET] i take it you've read that [02:43:21 CET] i've only ever built for windows with mingw [02:46:29 CET] furq, thank alot [02:46:43 CET] wannabegeekster, thank for your help [02:51:37 CET] furq, I am using MSVC 2015, is it OK? [03:22:24 CET] hi, seems to be a problem when using more than one -headers option, the second one is not included in the request [03:23:35 CET] The second one isn't included? I'd assume the first one wouldn't be included. [03:23:39 CET] What's your commandline? [03:24:19 CET] I added -report after and saw the request and only one of the headers was added [03:24:30 CET] Yes [03:24:41 CET] It's probably the second one though, not the first. [03:24:50 CET] Duplicate options overwrite each other.\ [03:25:09 CET] even if they dont contain the same stuff? [03:25:11 CET] If you want to specify multiple headers, separate them with crlf (I think \r\n should work too) [03:25:36 CET] ah, does it work in windows too? [03:25:51 CET] should [03:26:43 CET] X-Token: stuff\r\nX-Token2: stuff2 [03:26:48 CET] just output this [03:29:24 CET] btw, is it possible to remove the Icy-MetaData 1 header? Seems to be default on [03:31:30 CET] -icy 0 [03:31:39 CET] thanks [03:39:43 CET] Fredro: you'll have to find some way to actually encode CRLF into the string. [03:39:59 CET] The ascii values for the bytes, that is. [03:40:33 CET] ah ok, I'll give it a try [04:02:00 CET] Hi Furq, I just used the tutorial link, but the error is same "ERROR: libx264 not found" [04:02:24 CET] my environment: VS 2015, 64 bit [04:03:14 CET] x264 -> ./configure --enable-shared --enable-static --enable-win32thread --disable-thread --extra-cflags="-DNO_PREFIX" --prefix=/usr/, make, make install-lib-shared [04:03:29 CET] fmpeg -> ./configure --enable-asm --enable-yasm --arch=i386 --disable-ffserver --disable-avdevice --disable-swscale --disable-doc --disable-ffplay --disable-ffprobe --disable-ffmpeg --enable-shared --disable-static --disable-bzlib --disable-libopenjpeg --disable-iconv --disable-zlib --enable-libx264 --prefix=/ffmpeg/ --toolchain=msvc --extra-cflags="-MDd" --extra-ldflags="/NODEFAULTLIB:libcmt" --enable-debug [08:03:02 CET] wannabegeekster [08:03:07 CET] are you there? [08:04:23 CET] Furq, are you there? [08:25:34 CET] chaupv: you need --extra-cflags="-I/path/to/prefix/include" --extra-ldflags="-L/path/to/prefix/lib" [08:26:13 CET] where prefix is the one you used to install libx264 [08:27:09 CET] I have done it already, but the error still occurs... [08:27:28 CET] ./configure --enable-asm --enable-yasm --arch=i386 --disable-ffserver --disable-avdevice --disable-swscale --disable-doc --disable-ffplay --disable-ffprobe --disable-ffmpeg --enable-shared --disable-static --disable-bzlib --disable-libopenjpeg --disable-iconv --disable-zlib --enable-libx264 --enable-gpl --enable-pic --prefix=/ffmpeg/ --toolchain=msvc --extra-cflags="-MDd " --extra-cflags=-I/usr/include --extra-ldflags="/NODEFAULTLIB [08:27:49 CET] configure --enable-asm --enable-yasm --arch=i386 --disable-ffserver --disable-avdevice --disable-swscale --disable-doc --disable-ffplay --disable-ffprobe --disable-ffmpeg --enable-shared --disable-static --disable-bzlib --disable-libopenjpeg --disable-iconv --disable-zlib --enable-libx264 --enable-gpl --enable-pic --prefix=/ffmpeg/ --toolchain=msvc --extra-cflags="-MDd " --extra-cflags=-I/usr/include --extra-ldflags="/NODEFAULTLIB:li [08:28:42 CET] --extra-ldflags="-L/path/to/prefix/lib" [08:28:46 CET] -L [08:31:12 CET] yes, already did it, but no change... [08:31:51 CET] pastie.org the end of your config.log [08:32:11 CET] I compiled and make install x264 to /usr/ and check for libx264.lib, it has already been there.. but ffmpeg cannot find it [08:32:38 CET] LINK : warning LNK4044: unrecognized option '/Wl,'; ignored LINK : warning LNK4044: unrecognized option '/Bsymbolic'; ignored LINK : warning LNK4044: unrecognized option '/LC:/environtment/msys64/usr/lib/'; ignored LINK : fatal error LNK1181: cannot open input file 'x264.lib' ERROR: libx264 not found [08:33:48 CET] may be , it dont know this path with prefix '/LC:/?... [08:34:09 CET] it doesn't know.. [08:37:40 CET] you need to pastebin your whole configure line and the command in config.log [08:37:49 CET] last command* [08:39:15 CET] Ok [08:39:21 CET] here is my configure command [08:39:36 CET] ./configure --enable-asm --enable-yasm --arch=i386 --disable-ffserver --disable-avdevice --disable-swscale --disable-doc --disable-ffplay --disable-ffprobe --disable-ffmpeg --enable-shared --disable-static --disable-bzlib --disable-libopenjpeg --disable-iconv --disable-zlib --enable-libx264 --enable-gpl --enable-pic --prefix=/ffmpeg/ --toolchain=msvc --extra-cflags="-MDd " --extra-cflags=-I/usr/include --extra-ldflags="/NODEFAULTLIB [08:39:36 CET] ug [08:39:47 CET] fucking pastebin.com [08:39:56 CET] it's a website, go there [08:40:37 CET] Oh dear, sorry, [08:41:07 CET] and I said pastie.org earlier [08:41:22 CET] it doesn't matter which you use [08:41:44 CET] yes, I see [08:42:03 CET] this pastebin.com for configure [08:42:04 CET] http://pastebin.com/Yy93zkiF [08:43:22 CET] this for log [08:43:23 CET] http://pastebin.com/0XAn4Peg [08:44:11 CET] you only pasted the top of the log [08:44:25 CET] wait... [08:44:37 CET] log is too long, i am doing the remain.. [08:46:30 CET] http://pastebin.com/Q42JtFMX [08:47:24 CET] http://pastebin.com/MGVMq6kp [08:47:29 CET] Ok, man, [08:47:43 CET] I done the pastebin.com for log [08:48:27 CET] ok, but I think you still didn't get to the bottom [08:49:09 CET] becuase the last bit should show why libx264 was not found [08:51:30 CET] No, Man [08:51:43 CET] I copied all [08:52:38 CET] did you mean the last blank line? [08:53:27 CET] http://pastebin.com/TZW7myAs [08:53:55 CET] the last piece of log [08:54:39 CET] Ok, that's helpful :) [08:55:51 CET] :), relaxed, please help [08:56:28 CET] is that the latest version of visual studio? [08:56:47 CET] yes, VC 2015 [08:58:16 CET] did you read https://trac.ffmpeg.org/wiki/CompilationGuide/MSVC ? [08:59:49 CET] yes, several time, I seted up ffmpeg and x264 by using that guideline [09:00:09 CET] could you point out some tips? [09:00:24 CET] what's wrong with the configuration? [09:04:24 CET] I compiled ffmpeg without x264, but when i use --enable-libx264 --enable-gpl, the error occurs [09:06:40 CET] I've never compiled ffmpeg using MSVC. [09:07:03 CET] ha ha, me too [09:07:28 CET] In the log I see "C:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\BIN\cl.exe" but on MS's website it says the latest is 15 [09:08:31 CET] yes, it's VS 2015 [09:08:55 CET] I think that's no related to the problem... [09:09:05 CET] :) [09:10:30 CET] Relaxed, do you think it's better to replace VS version? should I use VS2013 instead? [09:12:33 CET] I think, it's the bug of ffmpeg when compile with VS2015 [09:22:33 CET] from looking at config.log I'd say it was your setup and not ffmpeg's fault [09:24:10 CET] ha ha... [09:24:49 CET] it's so difficult to figure out the error.. [09:42:12 CET] any tips, Relaxed? :) [10:00:31 CET] relaxed, where had you gone? [10:00:33 CET] ^^ [10:00:56 CET] you answer nothing for it, .. [10:04:43 CET] relaxed, you said like you are a geek for ffmepg but finally you have done nothing.. [10:17:11 CET] chaupv: I'm at work, it's not like I'm paid to sit in this channel to provide support. [10:19:04 CET] sometimes you have to wait in an irc channel until someone who knows what's going on looks at your issue. [10:19:48 CET] and I did help you get a helpful log for them to examine. [10:28:14 CET] relaxed: Ok, I got it :) [10:30:45 CET] chaupv: did you see the note in the wiki about cl and link.exe in the wiki? [10:31:20 CET] -"in the wiki" :) [10:31:57 CET] I don't know if that's the case, though [10:32:58 CET] I dont use link from Msys, I used MSVC link instead of it, also I use cl from VC [10:33:52 CET] I change the link.exe to link_backup.exe then I set environment that Msys can find MSVC link [10:34:51 CET] I think the cl, link ffmpeg wiki that you mentioned about does the same.. [10:35:54 CET] relaxed, how can i find you tomorrow? [10:36:17 CET] is it possible to find you in this chat room? [10:37:16 CET] Yes, I'm usually here 24/7 [10:37:51 CET] Oh, so you are a contributor for ffmpeg, [10:39:11 CET] I contribute time here to help people, small patches, and linux static builds [10:40:04 CET] Ok, you are power Man, :) [10:40:14 CET] I have a question? [10:40:44 CET] currently, I have a warning in the log like this: LINK : warning LNK4044: unrecognized option '/LC:/environtment/msys64/usr/lib'; ignored [10:41:24 CET] do you know what option to make link to become : "C:/environtment/msys64/usr/lib", without '/L'? [10:44:23 CET] in the wiki you're running ./configure options under "How to build debug version?". Why? [10:44:58 CET] try the simple "./configure --toolchain=msvc" first [10:45:31 CET] I have built successful ffmpeg without x246 [10:46:34 CET] I think if I can set link, compiler will find the libx246.lib [10:47:11 CET] Now, compiler doesn't know '/LC:/environtment/msys64/usr/lib' with prefix '/L' [10:49:30 CET] /L is the library search dir option [10:49:41 CET] so it should be /L "thing" [10:50:41 CET] is it valid for MSVC environment? [10:51:22 CET] I am using cl, link from MSVC, so I afraid of MSVC doesn'nt know 'L' [10:55:52 CET] relaxed? where prefix is the one you used to install libx264 ? -> it's /usr/ [10:59:13 CET] JEEB: here's his configure http://pastebin.com/0XAn4Peg [11:00:47 CET] and the bottom of his config.log http://pastebin.com/TZW7myAs [11:05:21 CET] relaxed, JEEB, thank for help, we will invoke this again tomorrow, I call it a day. hope to see you guys again tomorrow at this chat room. [11:07:00 CET] :) [13:41:57 CET] Do I need to use a filter to encode frames at vfr? [13:47:47 CET] dantti: no, you just need to have an input that's vfr [13:48:15 CET] c_14: ok but what then I set in time_base and pts? [13:48:52 CET] Just set the pts to the time you want the frame to be displayed at. [13:49:31 CET] Well, so that pts * time_base = time you want the frame to be displayed at [13:51:14 CET] but ain't time_base a fixed fraction ie 1, 25 = 25 fps? [13:51:41 CET] I guess then I need to store an initial clock, and then multiply that? [13:51:47 CET] time base is just a value that specifies on what scale PTS relate to seconds [13:52:25 CET] A timebase of 1000 (more precisely 1/1000) means that 1000 PTS = 1s [13:52:30 CET] so if 1, 25 time_base and I get 40 frames within a second is that a problem [13:52:36 CET] or should I drop frames [13:52:36 CET] So if you want a frame to display at 5s it would have the PTS 5000 [13:53:05 CET] dantti: yes, that's why the time_base is usually in the thousands at least [13:53:09 CET] the sourve is a live camera, that's why I'm saying I need to count the time [13:53:46 CET] Are you sure the source is vfr? [13:53:56 CET] c_14: so I guess there is no problem in jumping from 100pts to 1000 pts? [13:54:06 CET] If yes, then you need to count how many seconds have elapsed since you started, multiply that by the inverse of the time_base and use that as the PTS [13:54:18 CET] dantti: the time_base is usually constant for an entire video [13:54:20 CET] yup, just confirmed with the guy that implemented QCamera on android [13:54:39 CET] Just pick a time_base of 1000 [13:54:41 CET] It's usually fine [13:54:51 CET] ok [13:55:01 CET] That'll give you ms rounding though [13:55:10 CET] Ie your timestamps will be accurate to the nearest millisecond [13:55:28 CET] If you want a higher accuracy use a larger time_base [13:56:06 CET] (Note that I've occasionally been saying the timebase is 1000, but it's stored as 1/1000 in ffmpeg, it's just displayed as the inverse to make it easier to read) [13:57:27 CET] what does inverse means in math speaking (not a native speaker so I guess it's a math term which I don't get) [13:57:40 CET] 1000/1 I guess wouldn't be [13:57:40 CET] 1 / x [13:58:04 CET] 1/1000 [13:58:08 CET] yes [13:58:11 CET] ah ok [13:58:48 CET] thanks, let me try now to see if I got it right :) [14:04:01 CET] c_14: I just realized that QElapsedTimer can give me the time elapsed in milliseconds(since I call start() on first frame), so if I just set that to pts I guess I don't need to multiply to 1/1000 right? I can just use that value [14:04:51 CET] If your time_base is 1/1000, yes [14:06:40 CET] I think that didn't work, the file size is ~500kb and I just see a grey image [14:07:18 CET] on mpegts the log "nal 1" means what? [14:09:37 CET] That's just a TRACE log, not sure what it means though. [14:09:43 CET] is this a good value codecContext->bit_rate = 400000? [14:10:26 CET] from source av_log(s, AV_LOG_TRACE, "nal %d\n", state & 0x1f); I guess that's the mpegts state [14:10:30 CET] might be a bit low [14:12:15 CET] Action: c_14 needs to go now. Will be back later. [14:12:21 CET] ok thanks [14:51:36 CET] Hi guys, I want to convert an .MP4 movie to .AVI so that I can burn it as a DVD movie. [14:51:45 CET] I tried with this command, but audio is lost. [14:52:18 CET] ffmpeg -i movie.mp4 -codec copy movie.avi [14:56:54 CET] avi is equaly useless for a dvd as mp4. iirc the VOB stuff on DVDs is mpeg-ps. [14:57:14 CET] Hello all! I am attempting to use a filter that was recently developed and was told to apply the patch and rebuild ffmpeg. I cloned from source and compiled it but it says the filter was not found. Is there a place where I can download patches separately from the source repository? [14:59:35 CET] So I need to convert it to mpeg? [15:05:42 CET] iaskquestions: ffmpeg -i input -target ntsc-dvd output.mpg (or -target pal-dvd) [15:07:12 CET] iaskquestions: if it supports divx/xvid, that might be easier [15:08:06 CET] he already left the chat room [15:08:37 CET] that's what I get for ignoring parts/joins :/ [15:09:12 CET] Mo_YouVisit: which filter is it? [15:09:17 CET] panorama [15:09:29 CET] durandal_170: worked on it this past week [15:10:54 CET] find patch on ffmpeg-devel mailing list [15:15:11 CET] Mo_YouVisit: alternatively download my ffmpeg panorama branch from github.com/richardpl [16:29:11 CET] hey there - wondering if anyone on here knows how I could apply a blend filter and a fade filter at the same time. If I use the overlay filter instead of a blend, the fade works, but otherwise the blend is applied without the fade. Here's essentially what I'm doing right now: http://pastebin.com/Hp43NzS2 [16:38:55 CET] t4nk196: why not fade after blend? [17:10:33 CET] hi guys how to hide the dos console when calling ffmpeg trough a .bat file ? [ thanks ] [17:11:05 CET] is it normal for a 800x600 frame produce a pkt.size = ~204? when time_base was 1,25 and pts was -1 the pkt.size where around 1800, now that time_base is 1,1000 and pts is properly set the pkt size is quite small and the file doesn't has anything useful [17:11:25 CET] vircuser: this sounds like a windows problem, not an ffmpeg problem [17:11:49 CET] ok [17:12:26 CET] vircuser: I'd search for windows bat hide console [17:30:48 CET] hello guys [17:31:24 CET] hey .. can someone help me with vdpau ? [17:31:33 CET] i am running centos and i got vdpau compiled in [17:31:39 CET] keeps saying that i am unable to connect to X server [17:31:44 CET] what should i do to get decoding working ? [17:43:08 CET] since I get around 30fps setting using 1,30 as time_base and go increasing pts by 1, makes the encoded pkg.size ok, and the "first" frame is not a grey frame, but still the .ts file has a 0 duration even if I have frames for 6 seconds [17:46:13 CET] dantti: how are you getting the ts file duration? [17:46:38 CET] ts isn't an indexed format, it's rather a streaming format, so in general you can't get a duration value from it without reading the entire file. [17:46:49 CET] kepstin: when I open in any player it shows a frame and restarts [17:47:03 CET] not that it has a duration metadata [17:47:37 CET] on a valid ts vlc shows the duration because it can scan the file [17:48:55 CET] my code is this atm https://paste.kde.org/pg5ivvogv [17:49:40 CET] I can share the .ts file if it would help [17:59:26 CET] I have a feeling that I'm missing to set something in AvStream :P [18:08:07 CET] how should "-tune grain" command line look like with fmpeg/x265 ? [18:10:00 CET] ffmpeg seems happy about it [18:10:02 CET] http://paste.debian.net/plain/341223 [18:10:29 CET] but x265 [info]: tools: rd=3 psy-rd=0.50 rdoq=2 psy-rdoq=10.00 signhide tmvp < looks wrong [18:11:01 CET] according to this http://x265.readthedocs.org/en/default/presets.html#tunings [18:11:31 CET] rdoq should be 0, also rdoq level should be 0 [18:12:53 CET] unless iam reading all this worngly [18:12:58 CET] *wrongly [18:29:19 CET] for writing mpegts should I use av_interleaved_write_frame or av_write_frame? didn't unterstand it's difference [18:37:49 CET] Hello. I have a question on (I think) drawtext filter. I do live streaming with ffmpeg (kinda surveillance webcam) and put current time to the stream with "drawtext text='%{localtime}'". Is there a way to put a temperature (currently stored in a file) next to the date on the picture? When I use drawtext text='%{localtime} $(cat file-with-temp)' the temperature doesn't refresh on the file's changes. [18:43:03 CET] great now that I called av_packet_rescale_ts the video flows :) [18:51:57 CET] guys, is it possible to create a GIF file with ffmpeg? [18:51:58 CET] Hello. [18:52:09 CET] hoi [18:52:39 CET] I need to create quicktime movies with a larger-than-average audio chunk size. ffmpeg seems to like small chunks, whereas Apple products seem to like larger chunks, about 48k audio samples. [18:52:42 CET] Is there some way I can request that? [18:52:57 CET] Fyr: yes [18:53:14 CET] c_14, could you provide me with a link? [18:53:15 CET] There's the -chunk_duration option but I'm not sure what it refers to. [18:53:25 CET] Fyr: just use the gif output format? [18:53:50 CET] from a big number of PNGs into a GIF file. [18:54:08 CET] use the image2 input muxer [18:54:18 CET] ok [19:18:05 CET] hello, i cant find the proper way to get a palette for an animated gif, im trying this ffmpeg -v warning -ss 00:00:01 -i chichisg.mp4 -t 3 -vf "scale=320:-1:flags=lanczos,palettegen" -y palettechichis.png but nothing is done, i just want to get 3s from the start, what im i doing wrong, or how is the correct way to obtain a palette to be used as reference? [19:30:53 CET] hallo [19:31:08 CET] anyone alive here? [19:39:36 CET] Just before I ask question, little story. I'm writing ffmpeg gem for ruby and during development I notice something about fanky about avformat_close_input(). beside it's closing file it also frees memory of context [19:41:11 CET] so I'm working with context as class (instance) inside ruby and obviously when I close_file the DATA point to empty space in memory [19:42:04 CET] and ruby has it own free memory thing. and I'd like to deal with free on ruby side [19:43:18 CET] of course there is hack way to set pointer to NULL and check in every function if context == NULL, or just reallocate context [19:44:16 CET] Question1: Why does avformat_close_input() clear context [19:44:40 CET] Question2: is there way to close input without freeing context? [19:50:01 CET] A1: Context is related to input file. Closing input is reasonable: Context is filled with data from file and points to stream: It is file representation. [19:50:53 CET] A2: You could prealocate Context and extract Fields if you need data from context to persist after closing of file. [19:52:30 CET] I can see why context -> streams wont work, like pointed out it picking form file, but what about metadata and ect? what it be preloaded in moment of avformat_open_input? [19:53:09 CET] I have AVI file about which ffmpeg -i commands says mpeg4 (Advanced Simple Profile) (XVID / 0x44495658). And I need convert movies in MP4 to this mpeg4 (Advanced Simple Profile) (XVID / 0x44495658). How do I do this? What I am doing right now is using -vcodec libxvid option to convert video into mpeg4. [19:53:55 CET] Anyway thanks for answers, I probably will go with reallocating context after using avformat_close_input/_close_output [19:54:15 CET] But what about those profile? [19:54:16 CET] thought there was much clever way, that I was missing [19:58:26 CET] klaviatoora: you can copy it to mp4, try ffmpeg -i input -c copy output.mp4 [19:59:49 CET] sh4d0wD: Probably there is, but I seen many code where one just copied context data. See context as Object in OOP programming. [20:00:48 CET] What does -bf option mean? [20:01:47 CET] controls bframes [20:09:37 CET] y [20:12:05 CET] ented: ty. I actually got more exp. with C. Not big fun of OOP [21:36:31 CET] Hello, I just have a quick question about ffserver. Is it possible to pass on raw streams using ffserver? Or to enable minimal processing? Essentially i have a bunch of computers handling compression (with ffmpeg) and I'd like to have ffserver serve those compressed streams for me. [21:44:15 CET] hey running X centos.. configured vdpau for ffmpeg.. when it tries to connect it says it can't connect to X server .. can anyone help ? [21:45:06 CET] how do we duplicate the issue? [21:55:19 CET] llogan .. basically i am just trying to get vdpauinfo or vdpau working [21:55:26 CET] not sure how you duplicate : P [21:55:51 CET] sounds like you're asking the wrong channel [21:58:02 CET] its ffmpeg hwaccel option [21:58:14 CET] X server is on so.. i don't get it [21:58:36 CET] what command/code you are running [21:58:46 CET] can you pastebin it [21:58:54 CET] ffmpeg -hwaccel vdpau -i ... [21:59:07 CET] what is error message [21:59:12 CET] Cannot open the X11 display . [21:59:12 CET] VDPAU init failed for stream #0:9. [21:59:35 CET] what is argument for -i? [22:00:23 CET] what does: vdpauinfo | pastebint say? [22:01:36 CET] -i is udp://xxxxxx [22:02:07 CET] so ffmpeg -hwaccel vdpau -i udp://...... -f mpegts - | tee > ..... [22:02:13 CET] the error is jus tthe hwaccel [22:02:17 CET] says can't connect to X sever [22:02:24 CET] not sure why.. X server is running [22:02:31 CET] 22:01 < fritsch> what does: vdpauinfo | pastebint say? [22:03:00 CET] [root at enc4-5 scripts]# vdpauinfo [22:03:00 CET] vdpauinfo: cannot connect to X server [22:03:07 CET] tadaaaaa [22:03:09 CET] :-) [22:03:09 CET] that is all it says [22:03:13 CET] now fix your setup [22:03:18 CET] :-) [22:03:20 CET] lol that is the part i need help with :P haha [22:03:35 CET] you run as root - no idea why one runs such a command as root [22:03:45 CET] you run it with the user running X [22:03:50 CET] to access its session [22:03:50 CET] even as a user i get same error [22:04:02 CET] yeah - this is not a centos help channel [22:04:03 CET] but i dont' use the UI [22:04:03 CET] are you using shell login? [22:04:08 CET] its a server in a datacenter [22:04:12 CET] export DISPLAY=:0 [22:04:14 CET] .... [22:04:23 CET] or whatever ... running as root via remote shell [22:04:29 CET] no further comment [22:04:30 CET] ^^ [22:05:22 CET] of course i tried that but now that you said it :( [22:05:23 CET] [root at enc4-5 scripts]# vdpauinfo [22:05:23 CET] display: :0 screen: 0 [22:05:23 CET] Error creating VDPAU device: 1 [22:05:34 CET] at least i got something now . getting closer .. but still an error [22:06:26 CET] that video brand? [22:06:33 CET] nvidia [22:06:37 CET] M4000 card [22:06:43 CET] try export VDPAU_NVIDIA_DEBUG=3 [22:06:50 CET] and export VDPAU_TRACE=1 [22:07:04 CET] then pastebin here results of vdpauinfo [22:07:24 CET] i think there is not even the proper nvidia drivers installed [22:07:32 CET] ^^ [22:07:37 CET] see i am learning alrady :) thank you [22:07:44 CET] and lsmod [22:07:51 CET] http://pastebin.com/nRcwXHtC [22:08:13 CET] sh4d0wD: still want to fix a fucked up centos setup? [22:08:20 CET] http://pastebin.com/ZYVMfyHF [22:08:43 CET] nah just matter of interest, challenge solving problems [22:09:07 CET] sh4d0wD thank you for helping... [22:10:06 CET] ok 1 make sure use using nouveau or nvidia, they won't work togerther, uninstall one of them and blacklist [22:10:39 CET] ya i blacklisted the nouveau but let me just make sure its off [22:10:48 CET] although lsmod shows something [22:10:49 CET] it's not off and running [22:10:51 CET] :-) [22:11:00 CET] right... will try that right now :) [22:11:08 CET] boot with nomodeset [22:11:09 CET] 2. reinstall libvdpau [22:11:25 CET] and make sure you install right ach verstion [22:12:20 CET] and actually totally reinstall nvidia driver [22:12:28 CET] I think your libs are mixed now [22:12:43 CET] and stop running from root [22:12:44 CET] :) [22:17:16 CET] ok rebooting [22:17:33 CET] well i installed it from the repo so.. [22:17:50 CET] will try all those things in the next few min to make sure [22:19:18 CET] all will depend on what is you ld it may be mixed with open driver and nvidia one so uninstall one and reinstall other should help avoid doing relink by hand [22:25:02 CET] ok so .. done all the above.. still same [22:25:48 CET] wait.... wait...... [22:26:00 CET] maybe just maybe... [22:28:21 CET] OMG.... slow close i can taste it.. but... i reinstalled the nvidia driver and it noticed vdpau there so it said it had to install its libraries so that's good news [22:28:38 CET] only when i startx gives me an error now .. pain in the $@$!! [22:29:21 CET] now i get " Xlib: extension "GLX" missing on display ":1". " [22:29:24 CET] nvidia-xconfig [22:29:27 CET] so rebooting to see [22:29:33 CET] sudo nvidia-xconfig& [22:29:40 CET] sudo nvidia-xconfig * [22:29:49 CET] lol which is it ? haha [22:29:51 CET] without(*) [22:30:03 CET] reconfigure your X11 config [22:30:06 CET] to use nvidia [22:30:35 CET] got it.. ok.. i did do that before but it was when broken so let me try again... btw, this is great once i am done i wil understand.. and i will not forgot :P like a pro !! [22:30:39 CET] your xorg.conf probably pointing to mesa [22:31:50 CET] http://pastebin.com/pNuX8pJ4 [22:31:56 CET] never easy !! .. [22:32:06 CET] so i got that before too when running the nvidia-xconfig [22:32:15 CET] keeps giving me errors when that happens [22:35:45 CET] for one stop running as root [22:36:50 CET] 2 did you reboot after installing nvidia? [22:37:05 CET] yup rebooted [22:37:17 CET] for 1... let me su to a user [22:37:33 CET] lsmod | grep nvidia [00:00:00 CET] --- Sat Dec 5 2015 From burek021 at gmail.com Sat Dec 5 02:05:02 2015 From: burek021 at gmail.com (burek) Date: Sat, 5 Dec 2015 02:05:02 +0100 (CET) Subject: [Ffmpeg-devel-irc] ffmpeg-devel.log.20151204 Message-ID: <20151205010502.E1AE218A006B@apolo.teamnet.rs> [00:14:13 CET] ffmpeg 03Rostislav Pehlivanov 07master:ed08cbd7b172: aacenc_ltp: fix out of bounds memory access [00:36:08 CET] aac-pns-encode fails on kfreebds, last know commit is listed as N-76962-g736e2e2 [00:36:37 CET] seems it started failing some time ago, but I can't find any commit with a hash containing g736e2e2 [00:39:34 CET] The g is not part of the hash [00:40:44 CET] (It's hex after all) [00:40:52 CET] that g confuses many people. including me a few times [00:47:01 CET] one of ganesh's commits broke it, probably the tablegen utils and maybe changed bit allocation on that particular system [00:52:59 CET] what's the g mean anyway [00:55:26 CET] git [01:00:23 CET] ah [02:05:20 CET] ffmpeg 03Michael Niedermayer 07master:cafb19560401: avformat/utils: Fix "libavformat/utils.c:927:35: runtime error: signed integer overflow: -2450238577049583619 - 9223090561878065151 cannot be represented in type long long" [02:05:21 CET] ffmpeg 03Michael Niedermayer 07master:77a644e6fa4a: avcodec/h264_direct: Check CO-POCs before use [02:24:06 CET] nevcairiel: whenever I see the words nal, slice etc used in vp9 context, I get the creeps :D [02:38:07 CET] ffmpeg 03Sebastian Dr�ge 07master:eb3628d87f67: mpegtsenc: Fix off-by-one in indexing into Opus channel mapping arrays [02:38:08 CET] ffmpeg 03Vicente Olivert Riera 07master:a27401a05ba3: mips: rename mipsdspr1 to mipsdsp [03:19:30 CET] ffmpeg 03Michael Niedermayer 07master:9696a01f1262: avformat/mpjpegdec: Fix "libavformat/mpjpegdec.c:269:9: warning: passing argument 3 of av_stristart from incompatible pointer type" [03:52:48 CET] ffmpeg 03Yann Coupin 07master:d89c4846e0d6: avformat/movenc: HE-AAC correct FourCC in ISML [05:15:46 CET] HWAccelIntro is missing any info about QSV encoders/decoderes. Is there other place that shows it? [06:29:26 CET] Mista_D: there used to be a post on ffmpeg-devel about how to use it. cant comment on the usefulness though, and I gave up trying to install Intel's media sdk on linux a long time ago [06:30:06 CET] grep for a thread originated by Stefano Sabatini, I believe Ivan Uskov was the one who answered it [06:30:14 CET] might be a few months ago [09:48:19 CET] QSV is not a classic ffmpeg hwaccel. [10:28:07 CET] why are these ppl hijacking my bool thread [10:28:09 CET] :( [13:26:10 CET] Compn: drones are using equirectangular [13:32:34 CET] my single greatest fear! [13:32:35 CET] ehe [13:38:32 CET] ffmpeg 03Michael Niedermayer 07master:18268f761bff: avcodec/vp3: Fix "runtime error: left shift of negative value" [13:38:33 CET] ffmpeg 03Michael Niedermayer 07master:9d44e236187c: ffmpeg: Check pkt_dts for AV_NOPTS_VALUE [14:16:46 CET] ffmpeg 03Michael Niedermayer 07master:2ff61c3c1a0a: avcodec/dnxhddec: Fix runtime error: left shift of negative value [14:16:47 CET] ffmpeg 03Michael Niedermayer 07master:0c1d62ab9d75: avformat/oggparsetheora: Check duration to be not AV_NOPTS_VALUE (and positive) before use [14:55:35 CET] ffmpeg 03Hendrik Leppkes 07master:d6b988b505d6: libavutil/tablegen: add missing math.h include [14:55:36 CET] ffmpeg 03Hendrik Leppkes 07master:42a9aead854f: avcodec: include tablegen compat header before the tablegen functionality [14:55:37 CET] ffmpeg 03Hendrik Leppkes 07master:3c4e49187f83: libavcodec/aacsbr_tablegen: fix tablegen on windows [15:01:33 CET] BBB: is there a standard way of unpacking coefficients with multiple bit depths or is it a case of if(foo){AV_WN16}else{AV_WN32}? [15:01:50 CET] ah seems macroed in h264 [15:01:58 CET] I do that, with macros and inline funcs [15:02:09 CET] so the check is essentially once per coef block readoing, not once per coef [15:02:40 CET] yeah [15:02:45 CET] atomnuker: ^ [15:03:07 CET] youre probably going to say dude this is pretty hacky at some point, looking at how codecs implement multi-bitdepth [15:03:14 CET] and I cant entirely deny it [15:03:17 CET] can't any other way [15:03:19 CET] see* [15:03:22 CET] but it works and the performance impact is low, so I think its ok [15:03:25 CET] yeah exactly [15:03:33 CET] dude this is pretty hacky [15:04:34 CET] just write it in C++ and use templates [15:07:37 CET] c++ is unsafe, you should use python [15:08:32 CET] only pure functional languages are safe [15:08:55 CET] furthermore you should formally prove your code and algo is correct [15:09:02 CET] and publish 6 papers [15:13:30 CET] dont forget to get tenure [15:14:49 CET] that doesnt exist anymore [15:19:03 CET] python is soo slow [15:20:55 CET] ffmpeg 03Andreas Cadhalpun 07master:488e9a06f014: ffmdec: only seek back at EOF if a server is attached [15:20:56 CET] ffmpeg 03Andreas Cadhalpun 07master:a611375db532: ffmdec: reject zero-sized chunks [15:21:03 CET] durandal_170: thats only because your datacenter isnt big enough [15:21:13 CET] havent you heard of high performance computing? [15:21:14 CET] its very simple [15:21:18 CET] if your software is too slow [15:21:21 CET] just buy more datacenters [15:21:28 CET] and now your software is fast again!!!!! [15:21:41 CET] \o/ [15:22:54 CET] BBB, preferably using: FORTRAN, SciPy, OpenMP, and 900 nested loops [15:23:05 CET] yes, openmp [15:23:12 CET] and gccs vectorization [15:23:21 CET] I hear its really good at vectorizing memset(0) calls [15:23:24 CET] and thats about all it does [15:23:26 CET] double precision everywhere too [15:24:45 CET] speaking of gcc, 5.3 was released [15:25:31 CET] the only noteworthy highlight is skylake/avx-512 support, if you are into intrinsics [15:25:36 CET] speaking of papers, did you guys know I recently published a new paper? [15:25:48 CET] I think Im at 6 papers now, actually, so there you go Daemon404 [15:26:54 CET] ;) [15:28:03 CET] yeah I just counted, if I count only real published and peer-reviewed ones, its 6 papers [15:28:09 CET] if I include conference papers, its 8 [15:28:23 CET] eg the best papers ive read usually come from cons [15:28:31 CET] write papers, or write simd [15:28:32 CET] hm... [15:28:33 CET] difficult [15:28:46 CET] (siggraph) [15:52:31 CET] ffmpeg 03Cl�ment BSsch 07master:fb99ef0bd39a: avcodec: use AV_OPT_TYPE_BOOL in a bunch of places [15:52:32 CET] ffmpeg 03Cl�ment BSsch 07master:0e62b5d1ef13: avfilter: use AV_OPT_TYPE_BOOL in a bunch of places [15:52:33 CET] ffmpeg 03Cl�ment BSsch 07master:43ecec0f0386: avformat: use AV_OPT_TYPE_BOOL in a bunch of places [15:52:34 CET] ffmpeg 03Cl�ment BSsch 07master:c1f114a8c4ff: swresample: use AV_OPT_TYPE_BOOL for linear_interp and cheby options [15:52:35 CET] ffmpeg 03Cl�ment BSsch 07master:010caed261ac: swscale: use AV_OPT_TYPE_BOOL for {src,dst}_range options [15:52:36 CET] ffmpeg 03Cl�ment BSsch 07master:60532348d254: avcodec/mpegvideo: use constants for rc_strategy [15:52:37 CET] ffmpeg 03Cl�ment BSsch 07master:7234e04e358b: ffmpeg: enable echoing with command and debug modes [15:56:49 CET] hm gcc 5.3 tarballs not up yet [15:56:52 CET] lazy gnus [15:59:05 CET] are you sure its released and not gnu-released [15:59:14 CET] iirc gcc team had some weird defintiion of released [15:59:54 CET] they posted it on their main website [16:00:01 CET] GCC 5.3 released [2015-12-04] [16:01:39 CET] there is a tag and everything [16:01:41 CET] just no tarballs [16:02:30 CET] note: my fate instances should be re-up in the next 24 hours if everything goes well; sorry about the downtime [16:06:54 CET] "UndefinedBehaviorSanitizer gained a few new sanitization options" [16:06:56 CET] cool. [16:07:28 CET] note that this is the 5.0 changelog [16:07:35 CET] 5.3 changes are far on the bottom [16:08:02 CET] oh. [16:08:28 CET] :( [16:09:05 CET] michaelni: what's the usual way of testing sws? [16:09:21 CET] using vf_scale? [16:09:31 CET] fate [16:09:45 CET] yes well i mean within fate [16:10:03 CET] context: testing the nv/rgb convert [16:10:40 CET] not sure sws is that well tested in fate [16:10:59 CET] it has tons of tests [16:11:04 CET] all lavfi-fmt tests use sws [16:11:15 CET] but you could probably just use a simple ffmpeg command line and request a specific outout format and size [16:11:24 CET] but they take forever :-p [16:11:45 CET] have I ever mentioned I hate race conditions [16:11:53 CET] +1 [16:11:53 CET] I just had another one, SOOOOOOOOO frustrating to debug these things [16:12:23 CET] BBB: i spent a day writing a threadmessage fuzzer in order to make sure it wasn't broken [16:12:30 CET] (and it was, in a way) [16:12:37 CET] poor you [16:12:44 CET] and here the complexity is very very low [16:13:17 CET] btw, if you want to review that patchset& :) [16:13:32 CET] if the code is simple enough its usually easier to stare at it for a day than to try to reproduce it [16:17:25 CET] I don't know, I added a lot of threads to a very old single threaded codebase (mplayer), and the only kind I've run into was some missed wakeups [16:19:30 CET] BBB: so where is the race this time? [16:19:44 CET] not in ffmpeg :-p [16:19:46 CET] in my own code [16:19:48 CET] ahah [16:19:50 CET] ok [16:20:21 CET] closed source? [16:22:03 CET] not open, for now, sorry [16:22:19 CET] its that stream analyzer I showed at VDD [16:22:54 CET] its written using crazy mac tools anyway [16:22:56 CET] keep it [16:22:56 CET] :D [16:23:15 CET] and hide it [16:23:21 CET] :( [16:23:23 CET] everything for writing guis is pain [16:23:27 CET] he just chose what is least pain to him [16:24:19 CET] which reminds me i need to resurect the qp move to lavfi [16:26:49 CET] yes! [16:28:40 CET] isn't swift open source now [16:29:28 CET] (Its not GPL so no!) [16:30:18 CET] wm4, cocoa is not [16:30:32 CET] J_Darnley, surely the stdlib isnt [16:30:37 CET] or at least it must have an exceptions [16:30:38 CET] -s [16:31:20 CET] I just remember a link to a ballsy request on github to change the license to gpl [16:31:28 CET] that wasnt ballsy [16:31:32 CET] it was a lame troll [16:31:47 CET] If no asks it won't happen [16:32:07 CET] it was not a serious PR [16:54:49 CET] Action: Daemon404 wonders why mips is so interested in lavc [17:02:51 CET] they're too slow to run stuff there realtime? [17:03:30 CET] and they probably don't want to make their CPUs more performant [17:07:15 CET] BBB: done [17:11:53 CET] oh gcc tarballs showed up [17:14:59 CET] time to asess the damage [17:15:00 CET] ? [17:15:14 CET] its a bugfix release, unlikely to break much [17:15:36 CET] after i finally managed to figure out why my strip was randomly failing in mingw [17:16:34 CET] o ok [17:17:39 CET] hmm [17:17:53 CET] Action: Daemon404 is sure this options thread will in no way be controversial or massive [17:21:08 CET] "This is a rather long explanation on ideas " yeah, no [17:23:12 CET] just going over it without reading everything, there are probably good ideas in it, but still, changing Everything rarely works out too well [17:25:50 CET] wm4, a skim makes it seemed a little to aimed at lavfi [17:25:54 CET] too* [17:29:32 CET] he wants to make the parsers recursive, instead of embedding quoted strings [17:36:09 CET] I think his complaint about avoptions being designed primarily for ffmpeg.exe is correct [17:36:41 CET] i dont have any complains about the api myself [17:36:43 CET] our tools also have some local options btw [17:36:44 CET] i use it fairly extensively [17:36:53 CET] while we could maybe use AVOptions :p [17:37:18 CET] what i miss the most about AVOptions are lists of items [17:37:23 CET] :( [17:38:03 CET] using AVOptions from API is usually fine [17:38:29 CET] I think he doesnt mind the api [17:38:35 CET] I think he minds the string handling in the api [17:38:56 CET] the overabundance of strings, used for lists of devices when dealing with capturing video on mac/windows [17:39:06 CET] or used for all kind of other things [17:39:17 CET] a lot of that could be more clever [17:39:35 CET] clever is the death of simple [17:39:54 CET] and complex is the death of understandable [17:41:22 CET] word for sunday .. [17:41:34 CET] the places where the API wants lists are disgusting [17:41:48 CET] (though I doubt it would improve with these avoption changes) [17:59:14 CET] ffmpeg 03Paul B Mahol 07master:5d2cc00dd019: avfilter: add audio emphasis filter [18:01:37 CET] watch fate for me [18:09:27 CET] 30 band audio equalizer? [19:09:43 CET] ffmpeg 03Michael Niedermayer 07master:a85b02dcf70f: avcodec/jpeg2000dec: Check bpno in decode_cblk() [19:09:44 CET] ffmpeg 03Michael Niedermayer 07master:689a8674131c: avformat/msf: Check channels when reading the header [19:09:45 CET] ffmpeg 03Michael Niedermayer 07master:ab7ff38052f1: avformat/flvdec: Fix left shift of 137 by 24 places cannot be represented in type int [21:27:38 CET] ffmpeg 03Vicente Olivert Riera 07master:c8419035f48f: mips: remove --mips32r5 configure option [21:27:39 CET] ffmpeg 03Vicente Olivert Riera 07master:1a538a149063: mips: add p5600 core optimizations [22:30:24 CET] ffmpeg 03Michael Niedermayer 07master:4f2419888ba4: avcodec/pgssubdec: Fix left shift of 255 by 24 places cannot be represented in type int [22:30:25 CET] ffmpeg 03Michael Niedermayer 07master:325b59368dae: swscale/utils: Fix for runtime error: left shift of negative value -1 [00:00:00 CET] --- Sat Dec 5 2015 From burek021 at gmail.com Sun Dec 6 02:05:01 2015 From: burek021 at gmail.com (burek) Date: Sun, 6 Dec 2015 02:05:01 +0100 (CET) Subject: [Ffmpeg-devel-irc] ffmpeg.log.20151205 Message-ID: <20151206010502.7997F2AD6010@apolo.teamnet.rs> [00:25:06 CET] Hi all. On Debian, I'm using subsonic for media streaming. [00:25:30 CET] I'm trying to set it up to hardcode subtitles, with %s.srt (moviefile.srt) [00:25:38 CET] However, I'm getting a fontconf file error [00:25:50 CET] Example [00:25:51 CET] ffmpeg -ss %o -i %s -vf drawtext=fontfile=/usr/share/fonts/arial.ttf -vf subtitles=%s.srt -async 1 -b %bk -s %wx%h -ar 44100 -ac 2 -v 0 -f flv -vcodec libx264 -preset superfast -threads 2 -af volume=2.0 - [00:26:32 CET] [12/4/15 5:26:03 PM CST] INFO TranscodeInputStream Starting transcoder: [/var/subsonic/transcode/ffmpeg] [-ss] [644] [-i] [/media/zac/TORRENTS/DOWNLOADED/Apocalypto (2006) 1080p/Apocalypto (2006) 1080p.Molpol.mkv] [-vf] [drawtext=fontfile=/usr/share/fonts/arial.ttf] [-vf] [subtitles=/media/zac/TORRENTS/DOWNLOADED/Apocalypto (2006) 1080p/Apocalypto (2006) 1080p.Molpol.mkv.srt] [-async] [1] [-b] [2000k] [-s] [960x520] [-ar] [44100] [00:26:32 CET] [-ac] [2] [-v] [0] [-f] [flv] [-vcodec] [libx264] [-preset] [superfast] [-threads] [2] [-af] [volume=2.0] [-] [00:26:32 CET] [12/4/15 5:26:03 PM CST] INFO InputStreamReaderThread (/var/subsonic/transcode/ffmpeg) Fontconfig error: Cannot load default config file [00:26:40 CET] ooh, that's longer than I though, but yeah [00:30:39 CET] Does /etc/fonts/fonts.conf exist? [01:22:56 CET] c_14, I just made it, I don't get that error, but subtitles do not show in video [01:23:52 CET] [12/4/15 6:22:16 PM CST] INFO TranscodeInputStream Starting transcoder: [/var/subsonic/transcode/ffmpeg] [-ss] [233] [-i] [/media/zac/TORRENTS/DOWNLOADED/Apocalypto (2006) 1080p/Apocalypto (2006) 1080p.Molpol.mkv] [-vf] [drawtext=fontfile=/usr/share/fonts/arial.ttf] [-vf] [subtitles=/media/zac/TORRENTS/DOWNLOADED/Apocalypto (2006) 1080p/Apocalypto (2006) 1080p.Molpol.mkv.srt] [-async] [1] [-b] [2000k] [-s] [960x520] [-ar] [44100] [01:23:52 CET] [-ac] [2] [-v] [0] [-f] [flv] [-vcodec] [libx264] [-preset] [superfast] [-threads] [2] [-af] [volume=2.0] [-] [01:24:24 CET] Manually running a transcoding for 5 seconds, I don't get the hard coded fonts either [01:24:52 CET] This is the manual trans for webm, 5 seconds [01:24:53 CET] ffmpeg -ss 00:04:00 -i "/media/zac/TORRENTS/DOWNLOADED/Apocalypto (2006) 1080p/Apocalypto (2006) 1080p.Molpol.mkv" -t 00:00:05 -codec:v libvpx -quality good -cpu-used 0 -b:v 2000k -qmin 10 -qmax 42 -maxrate 1000k -bufsize 1000k -threads 4 -codec:a libvorbis -b:a 192k -vf "subtitles="/media/zac/TORRENTS/DOWNLOADED/Apocalypto (2006) 1080p/Apocalypto (2006) 1080p.Molpol.mkv.srt" st.webm [01:37:22 CET] This is the output. https://cluelessperson.com/st/st.webm [01:37:24 CET] no subtitles [01:39:19 CET] bbs, going home from office [01:41:54 CET] Good morning relaxed, JEEB, are you there? [02:39:19 CET] back [02:55:12 CET] Is there anyway to determine total frames and to encode the last frame of a video file? [03:02:29 CET] I found 1 method from stack overflow but its so slow. [03:32:25 CET] aphorise: what method? ffprobe with count_frames? [05:07:41 CET] relaxed, JEEB, are you there? [05:13:06 CET] hello, i was trying to put some subs to a movie, i tried this ffmpeg -i hannibal.102.hdtv-lol.mp4 -vf subtitles=hannibal.102.hdtv-lol.ass -vcodec h264 -acodec aac -scodec mov_text -metadata:s:s:0 language=spa -strict -2 -y hannibal-2.mp4 but it didnt work, when i try to see the file information it says [mov,mp4,m4a,3gp,3g2,mj2 @ 0x8368920] moov atom not found [05:13:11 CET] hannibal-2.mp4: Invalid data found when processing input [05:13:14 CET] what did i do wrong? [05:29:49 CET] hallo [05:55:20 CET] autofsckk: try ffmpeg -i hannibal.102.hdtv-lol.mp4 -i hannibal.102.hdtv-lol.ass -vcodec h264 -acodec aac -scodec mov_text -metadata:s:s:0 language=spa -strict -2 -y hannibal-2.mp4 [05:56:29 CET] the difference is to treat the subtitles as input instead of a filter? [05:56:49 CET] autofsckk: er, first try ffmpeg -i hannibal.102.hdtv-lol.mp4 -i hannibal.102.hdtv-lol.ass -map 0 -map 1 -c:v copy -c:a copy -scodec mov_text -metadata:s:s:0 language=spa hannibal-2.mp4 [05:57:37 CET] the subtitle filter will burn them into the video stream [05:57:43 CET] with c:v and c:a could it be a lot more faster? [05:57:50 CET] yes [05:58:37 CET] what if i want to do it for a dvd, i tried with -target ntsc-dvd but it changed the aspect of the video [05:59:15 CET] im making tests, im trying to learn how this works, its great really, but there are still too much things to learn/understand [06:00:55 CET] I would leave dvd/mpeg2 in the past where it belongs [06:01:49 CET] oh great, the second option did put the subs in place, i thought that only could be done with mkv files [06:02:51 CET] so instead of dvd, now how can i burn a movie with subtitles into a dvd to be played in a normal dvd player? [06:09:37 CET] does the dvd player support divx/xvid? [06:13:25 CET] im not sure, but i think it can [06:14:57 CET] so dvd/mpe2 is obsolete now? [06:15:58 CET] that would be much easier, ffmpeg -i hannibal.102.hdtv-lol.mp4 -vf subtitles=hannibal.102.hdtv-lol.ass -c:v mpeg4 -q:v 4 -vtag XVID -c:a libmp3lame -ac 2 -b:a 192k output.avi [06:17:18 CET] autofsckk: not obsolete but mpeg2 encoding can be slow and then you have to worry about anamorphic video and creating the dvd layout. [06:18:29 CET] whereas with divx/xvid video you can put the .avi on a cd or dvd and play it easily [06:20:00 CET] i know its super slow, i was testing with a 4G movie to convert ti to mpeg but ti was taking forever [06:20:34 CET] what does -q:v 4 do? [06:20:55 CET] it controls the video quality [06:21:42 CET] where can i read about that? formats? or coders? [06:22:07 CET] -qscale in the man page [06:23:01 CET] ok, ill look about it then, thanks [06:23:04 CET] one last question [06:23:11 CET] you should really use x264 if your player supports it [06:23:23 CET] it is both better and faster than xvid [06:23:43 CET] well yes, but do dvd players ever support h264? [06:23:53 CET] furq -> ok ill see if it supports it, i dont think so because its an old player [06:24:08 CET] "-vcodec h264" makes me think it does [06:24:11 CET] i have used movies with usb into new players [06:24:17 CET] if that was just an error then never mind [06:24:25 CET] also yeah use libx264 not h264 [06:25:14 CET] are those different? libx264 and h264? [06:25:21 CET] yes [06:25:36 CET] libx264 is a h264 encoder, so not really [06:25:50 CET] actually nvm i thought ffmpeg had an internal h264 encoder [06:26:51 CET] ah the question i had is that i have an mkv file and i would like to also put some subs on it, im seeing with ffmpeg -i file.mkv that it has chapters and all, is there a way to put it into a dvd and preserve that chapters stuff with the substitles too? [06:27:08 CET] i doubt it with ffmpeg [06:27:33 CET] there are dvd authoring tools which will do that [06:27:56 CET] but unless your player literally only supports mpeg2 then you should use something else [06:30:23 CET] i dont think it supports it, i went to see it and its really shitty jajajajaj svcd/vcd/cd [06:30:35 CET] its not mine, it belongs to my mother [06:31:00 CET] you should get her a chromecast for christmas [06:31:38 CET] (other brands of media player are available) [06:32:01 CET] i see [06:35:08 CET] i think the only way to go with that mkv file would be to turn it to svcd with the subs burned in it? [06:35:51 CET] you can convert it to dvd mpeg2 with burned-in subs [06:36:05 CET] you'll just need another tool if you want to create an actual video dvd [06:36:38 CET] like devede? [06:36:58 CET] sure [06:37:02 CET] i have to make first the mpeg2 with subs and then use devede right? [06:37:11 CET] that sounds about right [06:37:24 CET] will that process change the aspect ratio? [06:37:59 CET] not if the AR is 4:3 or 16:9 [06:39:06 CET] its wide, im not sure if is 4:3 or 16:9 but its wide [06:39:18 CET] i used -target ntsc-dvd when i was testing [06:39:32 CET] culd that be the reason why it changed the ratio? [06:43:21 CET] ntsc-dvd is always 720*480 [06:43:37 CET] you need to specify -aspect 16:9 or whatever [06:46:14 CET] ok, ill read more about that, thanks [06:47:26 CET] one thing i found very strange is that, well ffmpeg put my cpu to 100% and i have to use cpulimit, but i use it with limit to 100% and it still diminish the effect on it by 50%, its really strange that [06:56:52 CET] thanks for your help relaxed and furq good night :D [11:41:51 CET] Good morning (ugt) [11:42:27 CET] good morning [11:43:12 CET] any c developers alive here at this moment? [11:43:30 CET] yes? [11:43:34 CET] cool [11:44:50 CET] is there way to determent if AVFormatContext is still open [11:45:25 CET] in my code there 2 places where it can call avformat_close_input [11:45:54 CET] problem if I second call happens it sig faulting [11:46:12 CET] what happens? [11:46:31 CET] check if it is 0? [11:46:38 CET] one moment let me postebin it somewhere [11:48:19 CET] https://gist.github.com/sh4d0wPhoenix/95fd83be24904a537bb6 [11:48:49 CET] problem happen round line 18 in r_avformat_context_free [11:49:46 CET] and only happens if I close input before it call this free function [11:52:36 CET] and here is backtrace https://gist.github.com/sh4d0wPhoenix/7583090cda4e9a4636c0 [11:53:24 CET] if I don't use r_avformat_close_input_file [11:53:28 CET] works just fine [11:59:50 CET] in onxydocs it says about avformat_close_input_file that frees it and all its contents and set *s (pointers?) to NULL. [12:19:00 CET] nvm, solved it. it was on ruby side [12:19:30 CET] it was typecasting to what ever original pointer was [12:20:01 CET] but close_input make is nullptr, so there was memory mess [12:26:14 CET] Hi [12:26:56 CET] Can ffmpeg be recommended for DVD Video compliant MPEG-2 encoding under Linux? I know, there are probably better encoders under Windows... [12:33:28 CET] LigH: Don't see reason why it can't. ps there is no "better" encoder, at list for free and open [12:34:34 CET] but it all depends on what inside box and what load do you expect. [12:35:38 CET] OK, if you include "OpenSource" in the restraints, there is no choice. [12:35:55 CET] If you accept closed source, HCenc is one of the most recommended for Windows. [12:36:07 CET] But not for Linux. [12:36:35 CET] I'm just checking for a question of another forum user who used avconv and was surprised by a no-B-frame result. [12:36:47 CET] IP* only GOPs [12:37:18 CET] That's certainly not efficient. If compliant ... not really sure either, but probably? [12:39:44 CET] Hello! I'm trying to install ffmpeg git on Linux Mint, based on Ubuntu 14.04 LTS. [12:39:51 CET] Emmanuel_Chanel: hallo [12:40:24 CET] It says ERROR: x265 not found using pkg-config when I tried the installation in the way on https://trac.ffmpeg.org/wiki/CompilationGuide/Ubuntu [12:41:06 CET] sudo apt-get install libx264-dev [12:41:49 CET] Installing lib264 when missing libx265? [12:41:59 CET] fuck [12:42:02 CET] ops [12:42:08 CET] 4 am [12:42:11 CET] :) [12:42:13 CET] :) [12:42:31 CET] there is no libx265 in trusty [12:42:35 CET] I installed x265 but ffmpeg's ./configure doesn't detect. [12:43:52 CET] Maybe load the sources from bitbucket via mercurial, compile, and install.. [12:44:14 CET] Requires cmake [12:44:19 CET] why would that make a difference [12:44:42 CET] Independent of the distro. ;) [12:45:20 CET] is it independent of pkg-config [12:45:26 CET] because i assume that's the problem here [12:45:50 CET] >>I installed x265 but ffmpeg's ./configure doesn't detect. [12:47:02 CET] It says "x265 not found using pkg-config" [12:48:10 CET] https://github.com/videolan/x265/blob/master/source/x265.pc.in [12:48:12 CET] Ah, packaging is bureaucracy. :D [12:48:14 CET] it looks like it should create one [12:51:52 CET] Emmanuel_Chanel: pkg-config --variable pc_path pkg-config [12:52:00 CET] make sure x265.pc is in one of those paths [12:52:29 CET] Emmanuel_Chanel: did you try to use --extra-{c,cxx,ld}flags with configure and point to when-ever headers are.... or I'm missing problem and my brain just farted [12:52:59 CET] i mean you could do that, but that's what pkg-config is for [12:52:59 CET] Which installation? x265 or ffmpeg? [12:53:24 CET] ffmpeg [12:53:43 CET] where did you install x265 from [12:53:47 CET] source? [12:53:58 CET] source [12:54:10 CET] that should be ok then [12:54:36 CET] as long as it's not one of those shady PPAs i read so much about in the papers [12:55:23 CET] and super stupid question did u do make install when compiled x265? [12:56:26 CET] Yes. [12:58:43 CET] Emmanuel_Chanel: pastebin.com ffmpeg's config.log [13:02:02 CET] ok. [13:02:16 CET] Bye [13:04:37 CET] http://pastebin.com/MP2d5JnC [13:07:47 CET] Emmanuel_Chanel: it's x265's fault, the commited something that broke ffmpeg support. At the bottom, "/usr/local/include/x265.h:103:5: error: unknown type name 'bool'" [13:08:34 CET] add "#include " without the quotes to the installed x265.h [13:08:42 CET] then it will work [13:11:14 CET] Oh, it works... Thank you! I tried adding it to x265_config.h , it didn't work. So I couldn't expect you to be right. [13:11:43 CET] relaxed: Thank you very much! [13:12:04 CET] you're welcome [13:48:03 CET] I'm having difficulty with hardcoding subtitles [13:57:01 CET] I'm doing [13:57:04 CET] ffmpeg -ss 00:02:03 -i "Apocalypto (2006) 1080p.Molpol.mkv" -t 00:00:05 -codec:v libvpx -quality good -cpu-used 0 -b:v 2000k -qmin 10 -qmax 42 -maxrate 1000k -bufsize 1000k -threads 4 -codec:a libvorbis -b:a 192k -vf subtitles="Apocalypto (2006) 1080p.Molpol.mkv.srt" st.webm [13:57:09 CET] I see no errors [13:58:46 CET] https://cluelessperson.com/st/st.webm should have them [14:23:56 CET] cluelessperson: try with -ss after the input [14:27:09 CET] relaxed, that seems to take forever to do anything, what is it waiting on? [14:27:21 CET] oh, nvm [14:27:23 CET] watching [14:30:20 CET] relaxed, That works! :D [14:30:27 CET] relaxed, it seems slower to start though [14:32:33 CET] probably has to do with how seeking is done [14:39:05 CET] relaxed, Thank you, that works, but now the subtitles are wayyy off [14:39:08 CET] ffmpeg -i %s -ss %o -vf subtitles=%s.srt -c:v libx264 -preset superfast -b:v %bk -c:a libvorbis -f matroska -threads 2 -af volume=2.0 - [14:40:42 CET] Action: cluelessperson tries another subtitle file to be sure [15:01:34 CET] relaxed, Thank you again. https://cluelessperson.com/st/st.webm It works, fully [15:01:58 CET] Only thing, is that starting a video stream takes ~45 seconds, because the -ss AFTER input is SOO slow to seek [15:04:43 CET] Actually, much longer [15:04:56 CET] -ss after the input has to process/decode ALL of the video up to that point [15:05:02 CET] before skips ahead [15:09:56 CET] correct [15:10:13 CET] cluelessperson: use -ss before the input, but then use -copyts after the inputfile [15:10:40 CET] oh, I didn't think of that [15:10:55 CET] cbsrobot_, I'll try. I was just now tryin ffmpeg -ss something -i file -ss 0 -t 10 :D [15:11:03 CET] relaxed: I amended it to https://trac.ffmpeg.org/wiki/Seeking#Cuttingsmallsections [15:12:12 CET] ffmpeg --ss 00:05:00 -i file -ss 0 -t 10 does not work [15:12:15 CET] cbsrobot_, trying yours now [15:12:24 CET] I should add -start_at_zero to the wiki aswell - for mpegts etc. [15:13:42 CET] cbsrobot_, ffmpeg -ss 00:04:03 -i "Apocalypto (2006) 1080p.Molpol.mkv" -copyts -t 00:00:05 outputs an empty file [15:13:54 CET] Action: cluelessperson lookst at manpage [15:15:36 CET] cbsrobot_, I don't know hot to use -copyts could verify that for me? :) [15:16:48 CET] cbsrobot_, http://dpaste.com/0TD3QN4 :) [15:17:45 CET] I'm just overwriting this webm for testing. http://cluelessperson.com/st/st.webm [15:18:36 CET] hey, guys, does anybody here use Pymol? [15:21:24 CET] cluelessperson: probably it has some problems with cutting your file at I-frames [15:21:58 CET] cluelessperson: try -t 30 [15:22:09 CET] and see if it encodes 30 seconds [15:25:19 CET] cbsrobot_, nope, still nothing [15:26:11 CET] let me know if you want that console output as well [15:26:27 CET] cluelessperson: and without -copyts the file is not empty ? [15:26:41 CET] cbsrobot_, correct. [15:29:32 CET] cluelessperson: you're using an old version. Try http://johnvansickle.com/ffmpeg/ and see if it works [15:39:36 CET] relaxed, cbsrobot_ http://dpaste.com/0MY0H6M The console output changed slightly, but same result. With -copyts , no video output. [15:40:29 CET] hey, guys, could you check the following link and suggest what tool was used to create such a beatiful picture? [15:40:29 CET] http://s2.postimg.org/3tstg6vax/bulks.png [15:40:31 CET] cluelessperson: at the bottom: frame= 119 fps=7.7 q=0.0 Lsize= 1366kB time=00:00:05.04 [15:40:39 CET] looks like it encoded 119 frames [15:41:10 CET] Fyr: powerpoint :-) [15:41:25 CET] it looks too professional. =) [15:41:33 CET] not really [15:41:34 CET] Fyr: Inkscape maybe [15:41:44 CET] cbsrobot_: that's for the one without copyts, he didn't paste the output for copyts [15:42:11 CET] cbsrobot_, is Inkscape simple to a newbie? [15:42:24 CET] c_14: you are right - well cluelessperson paste the other one aswell then [15:42:31 CET] Fyr: sure [15:42:39 CET] Fyr: what OS do you use ? [15:42:39 CET] cool, thanks [15:42:46 CET] cbsrobot_, I apologize, I gave you the wrong output. (I tried with and without -copyts to prove the new version was working) [15:42:47 CET] corrected [15:42:48 CET] http://dpaste.com/2BYNF1C [15:43:00 CET] Inkscape is present on every platform. [15:44:42 CET] Fyr: sure [15:45:00 CET] cluelessperson: omit the -t option and see if it encodes anything [15:45:29 CET] cluelessperson: you can hit q to stop the encoding [15:46:11 CET] cbsrobot_, Seems to be encoding, yes, I get some weird messages I didn't see with the previous version (weird messages without copyts as well) [15:46:19 CET] cbsrobot_, I'm checking the webm and will paste the output [15:47:17 CET] cbsrobot_, https://cluelessperson.com/st/st.webm Okay, so really quick seeking, subtitles look good. getting console [15:48:54 CET] cbsrobot_, http://dpaste.com/2Y32P97 [15:49:02 CET] Those odd messages pop up [15:53:00 CET] cbsrobot_, Excellent! so you've solved my use case! :D but can we also get -t working as well? :) [15:54:19 CET] cluelessperson: not sure what is happening there [15:55:50 CET] cbsrobot_, You do bitcoin? [15:56:36 CET] cluelessperson: no [15:57:11 CET] cluelessperson: just help another person that need help and that is reward enough for me [15:57:50 CET] cbsrobot_, Want to learn bitcoin? XD [15:58:10 CET] cluelessperson: not atm, thanks [16:00:41 CET] cbsrobot_, Weirdly, this method seems to lose audio deep into a video [16:02:56 CET] >14 minutes <20 minutes [16:06:15 CET] Audio works as long as you leave it streaming past 16 minuts [16:06:26 CET] but around 16 minutes, if you seek further, audio cuts out. [16:16:47 CET] I have a 3 hour long video and towards the end the audio is out by about 7 seconds... however it is in sync at the start. [16:17:27 CET] is there a way I can stretch the audio over the 3 hours [16:23:22 CET] sine0: hallo [16:28:49 CET] hallo [16:29:04 CET] sine0: wie geht's ? [16:29:48 CET] cbsrobot_, Removing -async 1 fixes my audio cutting out [16:30:13 CET] cluelessperson: nice [16:47:52 CET] how can I turn subtitles off by default in an MKV container? none of them seem to be marked default [16:48:36 CET] oh, it must be just MPC turning them on, they aren't shown by default in mpv [17:00:36 CET] nyuszika7h, the application [22:00:07 CET] hey guys - I need to check if a file has audio in a bash script, and my googling is failing to come up with answers. Anyone know how I could do that? [22:03:11 CET] somethine using ffprobe like this maybe? ffprobe -i INPUT -show_streams -select_streams a -loglevel error [22:05:39 CET] seems like that should work [22:11:21 CET] thanks pzich, seems to work - returns nothing if there is no audio [22:11:28 CET] follow up question [22:12:27 CET] I'm concatenating with filter_complex, can I do that with a file that has no audio (while the rest do) or does it need audio? [22:14:11 CET] I know I can add silence if I have to, just wondering if I can avoid that step [22:16:09 CET] example of concat command: ffmpeg -i INPUTS -filter_complex [0:v]scale=320:-2,pad=320:480,setsar=sar=1/1[v0]; [1:v]scale=320:-2,pad=320:480,setsar=sar=1/1[v1]; [2:v]scale=320:-2,pad=320:480,setsar=sar=1/1[v2]; [3:v]scale=320:-2,pad=320:480,setsar=sar=1/1[v3]; [4:v]scale=320:-2,pad=320:480,setsar=sar=1/1[v4]; [v0] [0:a:0] [v1] [1:a:0] [v2] [2:a:0] [v3] [3:a:0] [v4] [4:a:0] concat=n=5:v=1:a=1 [v] [a] -map [v] -map [a] out [22:16:09 CET] put.mp4 [22:26:51 CET] I suspect you'll have to add silence, but I'm not 100% sure [22:27:51 CET] alright i'll test it. appreciate the help pzich! [23:38:20 CET] pzich you still around? have another nooby question for ya [23:39:19 CET] Getting this error from the concat command sometimes - I assume it's cause the input file has too large a resolution [23:39:23 CET] Input area 0:0:320:568 not within the padded area 0:0:320:480 or zero-sized [23:39:44 CET] I'd like to keep the aspect ratio and scale down to fit [23:40:05 CET] concat command: ffmpeg -i INPUTS -filter_complex [0:v]scale=320:-2,pad=320:480,setsar=sar=1/1[v0]; [1:v]scale=320:-2,pad=320:480,setsar=sar=1/1[v1]; [2:v]scale=320:-2,pad=320:480,setsar=sar=1/1[v2]; [3:v]scale=320:-2,pad=320:480,setsar=sar=1/1[v3]; [4:v]scale=320:-2,pad=320:480,setsar=sar=1/1[v4]; [v0] [0:a:0] [v1] [1:a:0] [v2] [2:a:0] [v3] [3:a:0] [v4] [4:a:0] concat=n=5:v=1:a=1 [v] [a] -map [v] -map [a] out.mp4 [23:54:05 CET] satt -> isnt w:h:x:y measures? [23:57:34 CET] autofsckk: I'm pretty new to this so I'm not sure exactly [23:58:06 CET] im new too jajajaja but i remember using crop filter and the structure is like that [23:58:45 CET] I think you can leave out the x/y positioning from the pad, since it autocenters [00:00:00 CET] --- Sun Dec 6 2015 From burek021 at gmail.com Sun Dec 6 02:05:03 2015 From: burek021 at gmail.com (burek) Date: Sun, 6 Dec 2015 02:05:03 +0100 (CET) Subject: [Ffmpeg-devel-irc] ffmpeg-devel.log.20151205 Message-ID: <20151206010503.8FF012AD6011@apolo.teamnet.rs> [00:37:25 CET] is there any way to turn off inlining in ffmpeg [00:42:19 CET] ubitux: there are multiple problems with 3e86ead3 [00:42:41 CET] ubitux: one being that your `while (i < FF_ARRAY_ELEMS(section->fields))` never increments i, so it loops forever [00:44:03 CET] ubitux: fortunately it never actually gets called, because its use is in an `else if()` after `if (section->format_header && !order)`, so in the no-format-header case, ass_split_section takes the previous branch and never makes it to the call [00:44:15 CET] I fixed both of those problems [01:14:56 CET] kierank: add __attribute__((noinline)) to the function in question [01:17:40 CET] I think there's an av_noinline alias for that [01:19:20 CET] yeah, you're right [01:51:28 CET] ffmpeg 03Andreas Cadhalpun 07master:7a4652dd5da0: aaccoder: prevent crash of anmr coder [02:18:21 CET] Daemon404, wm4, BBB, ubitux: I think you lot said earlier that you were using the API and AVOption. Can I ask where? I want to see what the use looks like from the outside. [02:26:13 CET] J_Darnley: well, there are 44 av_opt calls in mpv, most are in the libsw/avresample wrapper, and a libswscale wrapper, interestingly [02:27:14 CET] Thanks, I'll have a look. [02:27:56 CET] mpv.io right? [02:44:53 CET] iirc avresample/swscale/swresample are the only thing that ffms2 uses av_opt for too [02:57:23 CET] ffmpeg 03Claudio Freire 07master:293c170f5941: AAC encoder: ANMR, avoid empty search ranges [03:01:24 CET] do we actually need an XML parser [03:01:47 CET] that seems like a lot of work and a lot of complexity when there are at least 2 good libs for that [03:01:48 CET] Someone must do subs in that format! [03:02:02 CET] Or for some streaming format. [03:04:36 CET] don't joke about that stuff, there are ways to organize data and then there's xml and json [03:10:33 CET] Is it just javascript devs that think json is good (or decent)? [03:27:47 CET] json is actually easy to parse [03:27:54 CET] while xml is very hard [03:38:04 CET] hardest thing about parsing JSON is string escape handling, which is not very hard [04:48:53 CET] ffmpeg 03Simon Thelen 07master:5b6c0fdb4316: ffmpeg: When streamcopying, only add the input seek position when copying timestamps. [04:48:54 CET] ffmpeg 03Neil Birkbeck 07master:a16243a4aa5f: libavformat/mov.c: allow QuickTime metadata to come after traks [12:22:27 CET] ffmpeg 03Luca Barbato 07master:9f57f134c197: configure: ObjC support [12:22:28 CET] ffmpeg 03Hendrik Leppkes 07master:3b0f63e110d4: Merge commit '9f57f134c19773d54269b6cb9ee455ff87c2e9e1' [12:33:04 CET] "configure: ObjC support" << why? [12:33:45 CET] mac things, apparently [12:34:05 CET] (fwiw, we already have parts of that support before) [12:34:11 CET] s/have/had/ [12:36:20 CET] atomnuker: avfoundation [12:36:31 CET] we already have objc support since a while here [12:36:43 CET] ./libavdevice/qtkit.m [12:36:45 CET] ./libavdevice/avfoundation.m [12:36:54 CET] (that patch was totally broken, i had to fix quite a few things in it) [12:37:15 CET] that macbook actually came in handy for once [12:38:27 CET] ffmpeg 03Cl�ment BSsch 07master:560d1e7b4992: avfilter/codecview: add QP support [12:38:28 CET] ffmpeg 03Cl�ment BSsch 07master:3f46e7bad57c: avfilter/codecview: reindent after previous commit [13:40:29 CET] ffmpeg 03Michael Niedermayer 07master:7ed47e97297f: avformat/smacker: fix integer overflow with pts_inc [13:40:30 CET] ffmpeg 03Michael Niedermayer 07master:0c56f8303e67: avcodec/wmaprodec: Fix overflow of cutoff [13:40:31 CET] ffmpeg 03Michael Niedermayer 07master:2de8bfd2ef06: avcodec/pcm: Fix overflow in bitrate computation [13:40:32 CET] ffmpeg 03Michael Niedermayer 07master:15d14ce47cb3: avcodec/utils: Fix overflow in get_bit_rates computations [15:09:57 CET] hello! [15:10:08 CET] who's currently in charge with ffmpeg.org website? [15:13:28 CET] ubitux I believe [15:15:42 CET] i think you can sent patches against ffmpeg-web to the ML if you want to change something [15:17:55 CET] wat no! :( [15:18:24 CET] (no about maintainership) [15:21:19 CET] nevcairiel: sent that TrueHD-in-MPEGTS patch; apparently the sample plays in WMP with LAV Filters installed [15:21:41 CET] not sure if I should blame WMP or you [15:22:55 CET] ubitux: saw my reply re: ass_split? [15:23:27 CET] rcombs: no [15:23:34 CET] doesn't appear in the ml unless i missed sth [15:24:10 CET] on on irc my bad [15:24:17 CET] let me backlog, i missed a few messages [15:24:21 CET] yeah [15:25:12 CET] well I want to minimize the number of git:// users on git.videolan.org, so can someone please change the clone URI on http://ffmpeg.org/download.html#get-sources to http://source.ffmpeg.org/git/ffmpeg.git ? [15:25:43 CET] thresh: what's the difference? [15:25:46 CET] Action: kierank uses git.videolan [15:25:47 CET] rcombs: o damn, didn't realized that infinite loop; curious no static analyzer spotted it [15:25:59 CET] kierank: http:// is awesomer than git:// [15:26:08 CET] ubitux: maybe because it's never hit [15:26:08 CET] all the cool kids use it, just look at github [15:26:20 CET] pfft [15:26:31 CET] thresh: why http instead of https? [15:27:00 CET] rcombs: to hammer the server less, for no real reason? [15:27:22 CET] rcombs: it's curious bc i'm pretty sure i tested it against the problematic sample... but well alright, sorry [15:28:13 CET] ubitux: np as long as it gets fixed [15:28:24 CET] fate test welcome btw :( [15:28:49 CET] kierank: git:// is a binary proto, which is hard to mangle (which I need to migrate stuff from git.v.o to gitlab) [15:29:09 CET] I see [15:29:24 CET] ubitux: yeah, I'll put one together [15:29:35 CET] J_Darnley: i used the AVOption in a project recently; it was really handy to provide a way of doing mylib_set_option(ctx, "foo", ), with the mylib_set_option() last parem being vaarg doing introspection on type of the option to actually read the arg [15:29:48 CET] J_Darnley: rest was common usage [15:30:33 CET] it was actually really handy to set defaults value, range checks, and provide a stable option API [15:30:56 CET] like, not exposing the fields of the context and keeping it opaque, while shuffling everything in the background [15:41:42 CET] ffmpeg 03Michael Niedermayer 07master:66e05f6ff5e5: avcodec/wmaprodec: Check bits per sample to be within the range not causing integer overflows [15:48:29 CET] rcombs: wmp certainly doesnt support truehd i dont think, and not sure how I would modify truehd probing .. but maybe i forgot [15:48:43 CET] nevcairiel: it was a passthrough thing [15:48:54 CET] according to the user who provided the file [15:49:42 CET] the patch is trivial, so its probably fine [16:14:41 CET] Did they guy on ffmpeg-devel just post a pastebin link wrapped in a facebook tracking thing?! [16:15:00 CET] Also, on the wrong ml. [16:15:45 CET] ffmpeg 03Paul B Mahol 07master:a525b844d9e4: avfilter/af_stereotools: check s->length size [16:15:46 CET] BtbN: did you see the vaapi hevc-10bit/ vp9 code landing? [16:15:52 CET] yes [16:16:08 CET] did you understand their handling of p1010? [16:16:09 CET] waiting for the vp9 hwaccel hooks to be merged, then going to implement both [16:16:17 CET] people post things like "my problem" and they really expect people to help? [16:16:17 CET] Haven't looked too much at it yet [16:16:21 CET] hm, vaapi 10 bit, I wonder how surfaces are handled here [16:16:31 CET] wm4: ^^ that was my question [16:16:40 CET] they are converted internally to rgba32 [16:16:42 CET] last time I checked, surfaces magically changed their format depending on random conditions [16:16:47 CET] wat [16:16:47 CET] but I don't understand the code :-) [16:16:50 CET] that's fucked up [16:16:55 CET] http://lists.freedesktop.org/archives/libva/2015-December/003759.html [16:16:56 CET] here [16:17:14 CET] see - they need support for vaPutsurface [16:17:55 CET] the patch linked just adds 10 bit support to vaPutSurface? [16:17:59 CET] yes [16:18:23 CET] "Currently it converts P010 to RBGA32" [16:18:25 CET] he writes [16:18:39 CET] with a bit of phantasie I even see the 4 8bit planes [16:18:43 CET] asians, so verbose comments [16:19:06 CET] oh I see what you mean [16:19:09 CET] it looks very fucked [16:19:19 CET] http://lists.freedesktop.org/archives/libva/2015-December/003775.html <- emil has the same question [16:19:36 CET] we need that cause of vaPutSurface [16:19:46 CET] but i'd like to understand how it's done [16:21:18 CET] http://lists.freedesktop.org/archives/libva/2015-December/003766.html [16:21:20 CET] here the decoder [16:21:54 CET] #define I_P010 2, 2, 2, {I965_16BITS, I965_8BITS}, 3, { {PLANE_0, OFFSET_0}, {PLANE_1, OFFSET_0}, {PLANE_1, OFFSET_16} } [16:24:13 CET] Action: wm4 wonders what hw supports this [16:24:18 CET] Brexton [16:24:20 CET] the new atom [16:24:34 CET] Broxton* [16:24:36 CET] broxton [16:24:37 CET] yes [16:24:39 CET] no backwards compat? [16:24:50 CET] http://lists.freedesktop.org/archives/libva/2015-December/003756.html [16:24:51 CET] no [16:24:55 CET] gen9 only [16:24:57 CET] it seems [16:25:07 CET] gen9 is skylake as well, fwiw [16:25:12 CET] but decoder features may not be the same [16:25:16 CET] jep [16:25:27 CET] seems like I bought skylake really for nothing [16:25:28 CET] they do it like they did with braswell: broadwel gpu + skl hevc decoder [16:25:38 CET] broxton: skl gpu + kabi lake hevc decoder? [16:25:49 CET] wm4: haha [16:26:02 CET] yeah - windows + gpu assisted decoding [16:26:11 CET] but yes - codec wise totally not interesting [16:26:13 CET] broxton doesnt have a release date yet [16:26:14 CET] could have got a braswell [16:26:15 CET] so there is that [16:27:46 CET] i think the libyami guys will implement those bits the next days [16:27:55 CET] skylake always looked like a rather underwhelming chip, both on the cpu and gpu front, so.. [16:28:07 CET] i liked it concerning the 8 bit hevc [16:28:12 CET] think of broadwell ... [16:28:18 CET] it delivered absolutely nothing [16:28:29 CET] 8-bit hevc is not very interesting [16:28:47 CET] yes .. [16:28:54 CET] systems that adopt hevc for UHD appear to go 10-bit right away [16:29:01 CET] but never the less bsw / skl was used to implemented hevc vaapii nffmpeg [16:29:20 CET] and from the 10 bit hevc patch, it seems that extending "should work" quite well [16:29:23 CET] in vaapi side [16:29:42 CET] i got a braswell for vp9 hybrid decoding, but at least a braswell nuc is cheap, skylake cpus are expensive =p [16:29:52 CET] yeah [16:30:03 CET] how good is the performance of bsw vp9 decoding? [16:30:13 CET] 4k 24p possible? [16:30:18 CET] I don't think so [16:30:20 CET] didnt test [16:30:33 CET] you just implemented it and then did not test? :-) [16:30:45 CET] not the performance, cant really influence that anyway [16:30:55 CET] and ffmpegs cli dxva2 support is very inefficient [16:31:00 CET] so no good for benchmarking [16:32:35 CET] once its merged i might test it in my own code [16:33:22 CET] the bsw/skl hevc 8 bit decoder is really good [16:33:38 CET] and the bsw in case of decoding + rendering is also really nice [16:33:50 CET] 4k at 60 hevc 8 bit was doable [16:34:17 CET] therefore I am looking forward for these 10 bit chips - as they will decode uhd bluray and also new live tv standard [16:35:33 CET] just get a nvidia 950 or 960, they do it today =p [16:36:39 CET] yeah [16:36:41 CET] :-) [16:36:44 CET] not interested [16:36:53 CET] as the card needs 5 times what my bsw rig needs [16:36:54 CET] in idle [16:37:04 CET] its hard to be amazed by intel when they are 1-1.5 years behind the flow [16:37:12 CET] not sure they are [16:37:21 CET] at least on linux there is no 10 bit solution for now at all [16:37:33 CET] as multimedia devs you got to deal with broken consumer hardware and braindead consumers anyway [16:37:55 CET] hehe [16:41:23 CET] skylake should've had hevc 10-bit and hdmi 2.0, then it might've been worth-while of a platform [16:46:00 CET] jep [16:51:07 CET] skylake seems to be a disappointment in every way [16:51:14 CET] wait why? [16:51:14 CET] I guess haswell was too good? [16:51:27 CET] what does skylake have that is exciting? [16:51:32 CET] nothing [16:51:38 CET] unless you get wet from avx512 [16:51:40 CET] Ive been waiting for months now to get a new mbp until skylake gets in macs [16:51:51 CET] I think avx512 is nice [16:51:55 CET] and even then you need a xeon skylake [16:52:02 CET] never wait for new technology when buying something [16:52:22 CET] the current mac pros are & pre-sandybridge, I believe [16:52:47 CET] I have a broadwell mbp [16:53:03 CET] oh its ivy [16:53:06 CET] anyway [16:53:11 CET] ivy is not really exciting [16:53:17 CET] not even avx2 [16:53:54 CET] current mbps being sold are broadwell [16:54:11 CET] haswell no? [16:54:19 CET] or haswell, depending on which model [16:54:30 CET] 13" is broadwell, 15" is haswell [16:55:00 CET] but the haswell is slower [16:55:20 CET] the haswell is a quadcore, the broadwell only duals [16:55:24 CET] haswell clocks lower though [16:55:27 CET] so ... maybe? [16:55:43 CET] 2.2/2.5GHz quadcore haswell, vs. 2.7/2.9GHz dualcore broadwell [16:55:49 CET] which makes me frown [16:55:57 CET] there is a 2.8ghz haswell model [16:56:02 CET] but those upgrades costs a fortune [16:56:03 CET] ohright, yes [16:56:09 CET] 2.2/2.5/2.8 [16:56:12 CET] or 2.7/2.9/3.1 [16:56:18 CET] why not going with a broadwell thinkpad? [16:56:21 CET] I missed the highest [16:56:25 CET] he wants a mac =p [16:56:28 CET] core i7 broadwell? [16:56:29 CET] okay [16:56:40 CET] sorry mac only here, I dont have time to mess with my computer [16:56:52 CET] you do all the ffmpeg development on the mac? [16:56:55 CET] yes [16:57:55 CET] so why didnt they make a broadwell quadcore? so silly [16:58:19 CET] the broadwell only has a better gpu to my knowledge ... [16:58:24 CET] in comparison to hsw [16:58:53 CET] the interesting thing is that the haswell models were actually released after the broadwell 13" [16:59:02 CET] probably got a cheap deal on old haswell cpus =p [16:59:20 CET] or broadwell sucked heat / power wise? [16:59:31 CET] so maybe Ill just buy yhe mbp right now then [16:59:46 CET] I also want a mac pro, but paying $10k for a ivy seems kind of outrageous [17:00:47 CET] http://www.macrumors.com/roundup/macbook-pro/ <- some benchmarks [17:02:04 CET] no benchmarks 13 vs 15 though [17:02:39 CET] https://browser.primatelabs.com/mac-benchmarks [17:02:41 CET] perhaps there are some [17:03:18 CET] develops open source software ... uses proprietary kernel :P [17:03:19 CET] Action: Compn runs [17:03:50 CET] no system is entirely open source [17:04:15 CET] oh it it? nice [17:04:50 CET] since skl intel needs firmware [17:04:56 CET] too for the cpu [17:05:06 CET] or was that gpu part? [17:05:10 CET] gpu I think [17:07:23 CET] Compn: sorry Im over that period, I dont think opensource is the magic key to world domination. look at gnome or kde& really not that great [17:07:33 CET] im just kidding [17:07:41 CET] use what works. [17:08:44 CET] Compn: ;) [17:10:14 CET] Today is a slow day, and you people are so bored you only speak of hardware [17:10:17 CET] How about a nice big portion of IT'S HAPPENING [17:10:24 CET] ffmpeg 03Rostislav Pehlivanov 07master:3a6e0208615e: aacenc: mark the "faac"-like coder for removal [17:10:25 CET] ffmpeg 03Rostislav Pehlivanov 07master:b270ec9a1069: aacenc: mark coders other than twoloop as experimental [17:10:26 CET] ffmpeg 03Rostislav Pehlivanov 07master:e34e3619a2b5: doc/encoders.texi: update documentation for the native AAC encoder [17:10:27 CET] ffmpeg 03Rostislav Pehlivanov 07master:d9791a8656b5: aacenc: remove the experimental flag [17:10:30 CET] what happening ? aac [17:10:31 CET] oooo [17:10:35 CET] so pretty [17:10:53 CET] ffmpeg.org news entry? [17:10:58 CET] worthy. [17:11:38 CET] can we close the mega ticket now [17:11:59 CET] careful, the ticket is sentient now. [17:12:06 CET] and direct people to create new ones for new complaints =p [17:14:44 CET] not a bad idea [17:14:58 CET] who manages ffmpeg.org the website? [17:15:03 CET] :) [17:15:34 CET] Timothy_Gu ? [17:15:39 CET] thresh: I might be able to commit to it [17:15:55 CET] jsut send a patch [17:16:01 CET] many ppl cna write to it [17:16:06 CET] where to? [17:16:07 CET] (all devs?) [17:16:27 CET] nevermind [17:17:04 CET] meanwhile, the end is near http://pastie.org/pastes/10611337/text :( [17:17:43 CET] who has twitter access? [17:18:03 CET] Compn: ping any idea where I can get more cfhd samples [17:22:19 CET] kierank: https://gist.github.com/thresheek/257f8f97660c97e8c60b [17:24:17 CET] atomnuker: in the doc, you can use "auto", "true", "false" [17:24:25 CET] instead of "-1" "1" "0" [17:26:19 CET] thresh: can you send it in format-patch style [17:27:20 CET] ubitux: at least you know what hardware to upgrade to now :D [17:27:26 CET] kierank: hope this works: https://gist.github.com/thresheek/66f4dc38da19aa3bea4c [17:36:34 CET] someone will need to start a hydrogenaudio thread to rip the encoder apart with criticism [17:36:52 CET] ubitux: be glad it's not a ssd [17:37:21 CET] thresh: pushed [17:37:37 CET] kierank: thank you! [17:37:56 CET] oh shit, it's not the only place [17:41:21 CET] ffmpeg 03Rostislav Pehlivanov 07master:49b82bc9562e: doc/encoders.texi: remove forgotten mention of "experimental" from libvo-aacenc [17:42:20 CET] kierank: https://gist.github.com/thresheek/4f6285e59e1f731bfbbf [17:42:22 CET] need more coffee [17:44:06 CET] ffmpeg 03Michael Niedermayer 07master:79798f7c57b0: avcodec/dirac_parser: Fix potential overflows in pointer checks [17:44:07 CET] ffmpeg 03Michael Niedermayer 07master:c7d6ec947c05: avcodec/dirac_parser: Add basic validity checks for next_pu_offset and prev_pu_offset [17:44:08 CET] ffmpeg 03Michael Niedermayer 07master:a08681f1e614: avcodec/dirac_parser: Check that there is a previous PU before accessing it [17:44:26 CET] thresh: here to: https://trac.ffmpeg.org/wiki/CompilationGuide/Centos [17:45:08 CET] and several other wiki places [17:45:59 CET] thresh: pushed [17:46:16 CET] I don't have an account on trac.ffmpeg.org [18:08:03 CET] thresh, how long will git:// still be available/work for ffmpeg ? [18:15:39 CET] michaelni: forever [18:16:20 CET] we do not plan to remove it, ffmpeg and x264 will be there [18:17:50 CET] what was the point of switching to http then? [18:17:51 CET] ahh, ok, i was already a bit scared, but why do you change the default ? or rather how does it help to change the default if git:// support is left ? [18:18:14 CET] I want to minimize the future usage of git:// [18:18:20 CET] why? [18:18:25 CET] passing through more broken networks? [18:19:15 CET] bad experiences with git-daemon, mostly [18:21:22 CET] I would want to move git.videolan.org to some other server while preserving the URIs, and while fixing git:// protocol on the fly is possible (manipulating tcp stream), I would rather keep it to a minimum [18:25:41 CET] thresh, if it would help, we could probably move git to our box too, we already have ffmpeg-web git on it, but it would require a volunteer to setup & test it and also setup hooks and all, that is assuming these git:// issues dont affect us [18:26:04 CET] I see no point in doing that, we're happy to host you guys! [18:26:18 CET] ok [19:07:29 CET] I've written a news story: http://paste.debian.net/hidden/f01697f2/ [19:08:01 CET] any feedback before sending it over to the ML? [19:17:10 CET] I feel like it could use some rewording [19:17:18 CET] but I'm not a native speaker myself [19:22:34 CET] Update changelog [19:23:49 CET] append at the end of version ? [19:24:15 CET] wait, there's already "extensive native AAC encoder improvements" [19:24:55 CET] maybe I should replace that instead? [19:26:17 CET] yes [19:37:13 CET] ffmpeg 03Rostislav Pehlivanov 07master:713e67e5d873: Changelog: append experimental flag removal to the AAC encoder entry [19:37:14 CET] ffmpeg 03Rostislav Pehlivanov 07master:cc68ca0cab67: doc/encoders.texi: use words intead of numbers to describe option states [19:44:17 CET] ffmpeg 03Rostislav Pehlivanov 07master:3112501daf71: aacenc: fix aac_pred option triggering an error [20:10:26 CET] ubitux: thanks, I just saw your message (Re: avopt) [20:18:21 CET] ffmpeg 03Rostislav Pehlivanov 07master:dcbe8d8abc9b: aacenc_ltp: use an AR filter for LTP encoding as well [20:50:33 CET] ubitux: there is someone spamming #ffmpeg with rejoins of channel [20:50:50 CET] fixed [20:58:43 CET] thx [21:53:29 CET] ubitux: still wanting to do nlmeans? [21:54:16 CET] i'm currently working on it [21:54:41 CET] assom [23:07:59 CET] ffmpeg 03Michael Niedermayer 07master:214085852491: avcodec/hevc: Fix integer overflow of entry_point_offset [00:00:00 CET] --- Sun Dec 6 2015 From burek021 at gmail.com Mon Dec 7 02:05:03 2015 From: burek021 at gmail.com (burek) Date: Mon, 7 Dec 2015 02:05:03 +0100 (CET) Subject: [Ffmpeg-devel-irc] ffmpeg-devel.log.20151206 Message-ID: <20151207010503.8748B2AD601B@apolo.teamnet.rs> [00:45:26 CET] ffmpeg 03Rostislav Pehlivanov 07master:fcd9c6334a88: doc/encoders.texi: fix typo and clarify mpeg2_aac_low [01:44:16 CET] kierank: trac needs another reboot for Yet Another Kernel Update. could you do that when you get a chance whenever you're near the machine? [02:04:03 CET] If you are familiar with decibels you can also use dB: [02:04:03 CET] ffmpeg -i input.wav -af "volume=10dB" output.wav [02:04:09 CET] why "dB" case sensitive ? [02:04:20 CET] and then type "db" and get parse error. terrible [02:06:11 CET] Good. Enforce accurate and correct units! /s [02:26:14 CET] ffmpeg 03Michael Niedermayer 07release/2.5:cc80478463dd: avcodec/cabac_functions: Fix "left shift of negative value -31767" [02:26:15 CET] ffmpeg 03Michael Niedermayer 07release/2.5:d3de02d9d442: avcodec/cabac: Check initial cabac decoder state [02:26:16 CET] ffmpeg 03Michael Niedermayer 07release/2.5:c6b81e252e7a: avcodec/hevc: Check entry_point_offsets [02:26:17 CET] ffmpeg 03Michael Niedermayer 07release/2.5:551e18aa9205: avcodec/jpeg2000dwt: Check ndeclevels before calling dwt_decode*() [02:26:18 CET] ffmpeg 03Michael Niedermayer 07release/2.5:6bc2f5cc0b23: avcodec/jpeg2000dwt: Check ndeclevels before calling dwt_encode*() [02:26:19 CET] ffmpeg 03Michael Niedermayer 07release/2.5:f9d6d02fd1d6: avcodec/hevc_cabac: Fix multiple integer overflows [02:26:20 CET] ffmpeg 03Michael Niedermayer 07release/2.5:d8c0546ecdb0: avcodec/hevc: allocate entries unconditionally [02:26:21 CET] ffmpeg 03Michael Niedermayer 07release/2.5:2b9e849c3712: avcodec/vp3: Clear context on reinitialization failure [02:26:22 CET] ffmpeg 03Michael Niedermayer 07release/2.5:6e28d0d0475f: avcodec/utils: Use 64bit for aspect ratio calculation in avcodec_string() [02:26:23 CET] ffmpeg 03Michael Niedermayer 07release/2.5:e5d162ae9bd5: avcodec/utils: Clear dimensions in ff_get_buffer() on failure [02:26:24 CET] ffmpeg 03Michael Niedermayer 07release/2.5:439452a851f0: avformat/dump: Fix integer overflow in av_dump_format() [02:26:25 CET] ffmpeg 03Michael Niedermayer 07release/2.5:60369668ca34: avutil/integer: Fix av_mod_i() with negative dividend [02:26:26 CET] ffmpeg 03Michael Niedermayer 07release/2.5:582a6035bc8d: avutil/mathematics: Do not treat INT64_MIN as positive in av_rescale_rnd [02:26:27 CET] ffmpeg 03Michael Niedermayer 07release/2.5:73870147f855: avcodec/mpeg4videodec: Check available data before reading custom matrix [02:26:28 CET] ffmpeg 03Michael Niedermayer 07release/2.5:9b28bbb10c6b: avcodec/vp3: always set pix_fmt in theora_decode_header() [02:26:29 CET] ffmpeg 03Michael Niedermayer 07release/2.5:eff24a1f13b1: avcodec/apedec: Check length in long_filter_high_3800() [02:26:30 CET] ffmpeg 03Michael Niedermayer 07release/2.5:238c7c8276e5: avutil/mathematics: return INT64_MIN (=AV_NOPTS_VALUE) from av_rescale_rnd() for overflows [02:26:31 CET] ffmpeg 03Michael Niedermayer 07release/2.5:c8ca4b327542: avutil/timecode: Fix fps check [02:26:32 CET] ffmpeg 03Timo Ter�s 07release/2.5:cfd1bc581099: mpegencts: Fix overflow in cbr mode period calculations [02:26:33 CET] ffmpeg 03Michael Niedermayer 07release/2.5:f41c1d3bfeca: avcodec/vp3: Fix "runtime error: left shift of negative value" [02:26:34 CET] ffmpeg 03Michael Niedermayer 07release/2.5:c108dfc6ccd8: avformat/smacker: fix integer overflow with pts_inc [02:26:35 CET] ffmpeg 03Michael Niedermayer 07release/2.5:4b77e0a33ee7: avcodec/wmaprodec: Fix overflow of cutoff [02:26:36 CET] ffmpeg 03Michael Niedermayer 07release/2.5:38c1ab17ea7e: avcodec/wmaprodec: Check bits per sample to be within the range not causing integer overflows [02:26:37 CET] ffmpeg 03Michael Niedermayer 07release/2.5:752a6591b2db: avcodec/dirac_parser: Fix potential overflows in pointer checks [02:26:38 CET] ffmpeg 03Michael Niedermayer 07release/2.5:87dbcf7cd28e: avcodec/dirac_parser: Add basic validity checks for next_pu_offset and prev_pu_offset [02:26:39 CET] ffmpeg 03Michael Niedermayer 07release/2.5:9ce8308ee364: avcodec/dirac_parser: Check that there is a previous PU before accessing it [02:26:40 CET] ffmpeg 03Michael Niedermayer 07release/2.5:5c0be549efc2: avcodec/hevc: Fix integer overflow of entry_point_offset [02:26:41 CET] ffmpeg 03Michael Niedermayer 07release/2.5:5b41c961469a: swscale/utils: Fix for runtime error: left shift of negative value -1 [02:26:42 CET] ffmpeg 03Michael Niedermayer 07release/2.5:e4f90cbd0dd0: avcodec/pgssubdec: Fix left shift of 255 by 24 places cannot be represented in type int [02:26:43 CET] ffmpeg 03Michael Niedermayer 07release/2.5:f217aca5ba79: avcodec/jpeg2000dec: Check bpno in decode_cblk() [02:26:44 CET] ffmpeg 03Michael Niedermayer 07release/2.5:4a91e424dae8: avcodec/vp3: ensure header is parsed successfully before tables [02:26:45 CET] ffmpeg 03Michael Niedermayer 07release/2.5:f0ae57aca4c9: avcodec/hevc: Check max ctb addresses for WPP [04:41:30 CET] ffmpeg 03Michael Niedermayer 07release/2.8:31e54f41a023: avcodec/hevc: Check entry_point_offsets [04:41:31 CET] ffmpeg 03Michael Niedermayer 07release/2.8:9d5bdca3d8af: avcodec/jpeg2000dwt: Check ndeclevels before calling dwt_decode*() [04:41:32 CET] ffmpeg 03Michael Niedermayer 07release/2.8:9ce96a688b18: avcodec/jpeg2000dwt: Check ndeclevels before calling dwt_encode*() [04:41:33 CET] ffmpeg 03Michael Niedermayer 07release/2.8:e34876954429: avcodec/hevc_cabac: Fix multiple integer overflows [04:41:34 CET] ffmpeg 03Michael Niedermayer 07release/2.8:1cbc2cb5084e: avcodec/hevc: allocate entries unconditionally [04:41:35 CET] ffmpeg 03Michael Niedermayer 07release/2.8:d32c9723a626: avcodec/vp3: Clear context on reinitialization failure [04:41:36 CET] ffmpeg 03Michael Niedermayer 07release/2.8:ce6dd54711b3: avcodec/hevc: Check max ctb addresses for WPP [04:41:37 CET] ffmpeg 03Michael Niedermayer 07release/2.8:fe89682d7c61: avcodec/utils: Use 64bit for aspect ratio calculation in avcodec_string() [04:41:38 CET] ffmpeg 03Michael Niedermayer 07release/2.8:7372b42b184d: avcodec/utils: Clear dimensions in ff_get_buffer() on failure [04:41:39 CET] ffmpeg 03Michael Niedermayer 07release/2.8:2fbf72358517: avcodec/h264_refs: Check that long references match before use [04:41:40 CET] ffmpeg 03Michael Niedermayer 07release/2.8:3d69716baefd: avformat/dump: Fix integer overflow in av_dump_format() [04:41:41 CET] ffmpeg 03Michael Niedermayer 07release/2.8:4d9999705f40: avutil/integer: Fix av_mod_i() with negative dividend [04:41:42 CET] ffmpeg 03Michael Niedermayer 07release/2.8:460710500e17: avutil/mathematics: Do not treat INT64_MIN as positive in av_rescale_rnd [04:41:43 CET] ffmpeg 03Michael Niedermayer 07release/2.8:8cd3def81d11: avcodec/mpeg4videodec: Check available data before reading custom matrix [04:41:44 CET] ffmpeg 03Michael Niedermayer 07release/2.8:aab65146afd8: avcodec/vp3: always set pix_fmt in theora_decode_header() [04:41:45 CET] ffmpeg 03Michael Niedermayer 07release/2.8:d295ddffe138: avcodec/apedec: Check length in long_filter_high_3800() [04:41:46 CET] ffmpeg 03Michael Niedermayer 07release/2.8:e04b039b1528: avutil/mathematics: return INT64_MIN (=AV_NOPTS_VALUE) from av_rescale_rnd() for overflows [04:41:47 CET] ffmpeg 03Michael Niedermayer 07release/2.8:f2258e98991c: avutil/timecode: Fix fps check [04:41:48 CET] ffmpeg 03Timo Ter�s 07release/2.8:aa9ac199b8c7: mpegencts: Fix overflow in cbr mode period calculations [04:41:49 CET] ffmpeg 03Michael Niedermayer 07release/2.8:b8621a2e9860: avformat/riffdec: Initialize bitrate [04:41:50 CET] ffmpeg 03Michael Niedermayer 07release/2.8:73966b01618e: avcodec/vp3: Fix "runtime error: left shift of negative value" [04:41:51 CET] ffmpeg 03Michael Niedermayer 07release/2.8:0e3ec7db5334: avformat/smacker: fix integer overflow with pts_inc [04:41:52 CET] ffmpeg 03Michael Niedermayer 07release/2.8:16c5da92d997: avcodec/wmaprodec: Fix overflow of cutoff [04:41:53 CET] ffmpeg 03Michael Niedermayer 07release/2.8:ded0a0415377: avcodec/wmaprodec: Check bits per sample to be within the range not causing integer overflows [04:41:54 CET] ffmpeg 03Michael Niedermayer 07release/2.8:b253035ab23d: avcodec/dirac_parser: Fix potential overflows in pointer checks [04:41:55 CET] ffmpeg 03Michael Niedermayer 07release/2.8:cf79fd0317ed: avcodec/dirac_parser: Add basic validity checks for next_pu_offset and prev_pu_offset [04:41:56 CET] ffmpeg 03Michael Niedermayer 07release/2.8:0fa92fee4338: avcodec/dirac_parser: Check that there is a previous PU before accessing it [04:41:57 CET] ffmpeg 03Michael Niedermayer 07release/2.8:10fc3d690c8b: avcodec/hevc: Fix integer overflow of entry_point_offset [04:41:58 CET] ffmpeg 03Michael Niedermayer 07release/2.8:af4454561338: swscale/utils: Fix for runtime error: left shift of negative value -1 [04:41:59 CET] ffmpeg 03Michael Niedermayer 07release/2.8:e5a404477153: avcodec/pgssubdec: Fix left shift of 255 by 24 places cannot be represented in type int [04:42:00 CET] ffmpeg 03Michael Niedermayer 07release/2.8:ea4d9cb3e559: avcodec/jpeg2000dec: Check bpno in decode_cblk() [04:42:01 CET] ffmpeg 03Michael Niedermayer 07release/2.8:d27d59fef7e3: avcodec/vp3: ensure header is parsed successfully before tables [04:42:02 CET] ffmpeg 03James Zern 07release/2.8:cb4985ea9aa2: libvpxenc: remove some unused ctrl id mappings [06:42:17 CET] What's so difficult about translating the OpenHEVC x86 intrinsics to assembly? [10:03:01 CET] ffmpeg 03Reynaldo H. Verdejo Pinochet 07master:21c34cb26154: libavutil: add version component accessor macros [10:10:09 CET] ffmpeg 03Paul B Mahol 07master:3e1724baf8aa: avfilter/af_stereotools: fix logic fail [10:52:48 CET] ffmpeg 03Michael Niedermayer 07release/2.8:913c642c21dd: Update for 2.8.4 [11:36:08 CET] new nvidia encoder sdk was released, the encode api header is now licensed MIT. :) [12:57:31 CET] ffmpeg 03Michael Niedermayer 07release/2.4:ab79e3d1a5f9: avcodec/libopusenc: Fix infinite loop on flushing after 0 input [12:57:32 CET] ffmpeg 03Michael Niedermayer 07release/2.4:6f08086992c3: avcodec/truemotion1: Check for even width [12:57:33 CET] ffmpeg 03Michael Niedermayer 07release/2.4:e5c9396a0293: avformat/dump: Fix integer overflow in aspect ratio calculation [12:57:34 CET] ffmpeg 03Michael Niedermayer 07release/2.4:7ffe708297e9: avutil/common: Document FFABS() corner case [12:57:35 CET] ffmpeg 03Michael Niedermayer 07release/2.4:c09fe0346d94: avutil/common: Add FFNABS() [12:57:36 CET] ffmpeg 03Michael Niedermayer 07release/2.4:79b16c6e5e2f: avformat/mov: Fix integer overflow in FFABS [12:57:37 CET] ffmpeg 03Michael Niedermayer 07release/2.4:f66787d3452c: swresample/swresample: Fix integer overflow in seed calculation [12:57:38 CET] ffmpeg 03Michael Niedermayer 07release/2.4:ec35bb729cdd: avcodec/tta: Check init_get_bits8() for failure [12:57:39 CET] ffmpeg 03Michael Niedermayer 07release/2.4:45c30d84b6c5: avcodec/svq1dec: Check init_get_bits8() for failure [12:57:40 CET] ffmpeg 03Michael Niedermayer 07release/2.4:47b6ea314df6: avcodec/g2meet: Fix potential overflow in tile dimensions check [12:57:41 CET] ffmpeg 03Michael Niedermayer 07release/2.4:873ee14b560f: avcodec/g2meet: Also clear tile dimensions on header_fail [12:57:42 CET] ffmpeg 03Ganesh Ajjanagadde 07release/2.4:3ea20e60dc5d: avfilter/af_asyncts: use llabs for int64_t [12:57:43 CET] ffmpeg 03Michael Niedermayer 07release/2.4:7cdd319b01cf: avcodec/mjpegdec: Fix decoding RGBA RCT LJPEG [12:57:44 CET] ffmpeg 03Simon Thelen 07release/2.4:c6769b6d56be: lavf/webvttenc: Require webvtt file to contain exactly one WebVTT stream. [12:57:45 CET] ffmpeg 03Michael Niedermayer 07release/2.4:b5b29b22c0f8: avcodec/rangecoder: Check e [12:57:46 CET] ffmpeg 03Michael Niedermayer 07release/2.4:ac19d8eb3ac7: avcodec/ffv1dec: Explicitly check read_quant_table() return value [12:57:47 CET] ffmpeg 03Michael Niedermayer 07release/2.4:a2a93b0a8f5d: avcodec/ffv1dec: Fix off by 1 error in quant_table_count check [12:57:48 CET] ffmpeg 03Michael Niedermayer 07release/2.4:71fc26403f14: avcodec/x86/sbrdsp: Fix using uninitialized upper 32bit of noise [12:57:49 CET] ffmpeg 03Andrey Utkin 07release/2.4:eaf03fa8304c: avformat/httpauth: Add space after commas in HTTP/RTSP auth header [12:57:50 CET] ffmpeg 03Michael Niedermayer 07release/2.4:a3753ba10df1: avcodec/ffv1dec: Clear slice coordinates if they are invalid or slice header decoding fails for other reasons [12:57:51 CET] ffmpeg 03Michael Niedermayer 07release/2.4:f0db50a57f7b: avcodec/ffv1dec: update progress in case of broken pointer chains [12:57:52 CET] ffmpeg 03Michael Niedermayer 07release/2.4:99dd039d743d: avcodec/ffv1: Initialize vlc_state on allocation [12:57:53 CET] ffmpeg 03Tobias Rapp 07release/2.4:1bb7529ac2ed: avutil/file_open: avoid file handle inheritance on Windows [12:57:54 CET] ffmpeg 03Michael Niedermayer 07release/2.4:2f89546333b5: avcodec/mjpegdec: Check index in ljpeg_decode_yuv_scan() before using it [12:57:55 CET] ffmpeg 03Michael Niedermayer 07release/2.4:5d9bee34f9ae: avcodec/mjpegdec: Reinitialize IDCT on BPP changes [12:57:56 CET] ffmpeg 03Michael Niedermayer 07release/2.4:fe191124a9ca: avcodec/ffv1dec: Check for 0 quant tables [12:57:57 CET] ffmpeg 03Michael Niedermayer 07release/2.4:51ac1ce9668f: libavutil/channel_layout: Check strtol*() for failure [12:57:58 CET] ffmpeg 03Michael Niedermayer 07release/2.4:a3ff74c872ab: avformat/xmv: factor return check out of if/else [12:57:59 CET] ffmpeg 03Michael Niedermayer 07release/2.4:76950e524733: avformat/xmv: Discard remainder of packet on error [12:58:00 CET] ffmpeg 03Michael Niedermayer 07release/2.4:7ce70e6914bf: avcodec/dirac_parser: Fix undefined memcpy() use [12:58:01 CET] ffmpeg 03Michael Niedermayer 07release/2.4:c15f5068ca83: avcodec/microdvddec: Check for string end in 'P' case [12:58:02 CET] ffmpeg 03Michael Niedermayer 07release/2.4:d73a8ae70f15: avcodec/jpeg2000dec: Clip all tile coordinates [12:58:03 CET] ffmpeg 03Michael Niedermayer 07release/2.4:ac302efb9151: avcodec/jpeg2000dec: Check for duplicate SIZ marker [12:58:04 CET] ffmpeg 03Michael Niedermayer 07release/2.4:00dc345a83b8: avcodec/utils: Better check for channels in av_get_audio_frame_duration() [12:58:05 CET] ffmpeg 03Michael Niedermayer 07release/2.4:f68ff799eb00: avcodec/ivi: Check image dimensions [12:58:06 CET] ffmpeg 03Michael Niedermayer 07release/2.4:52d332b044eb: avcodec/flashsv: Check size before updating it [12:58:07 CET] ffmpeg 03Michael Niedermayer 07release/2.4:99e080ec38ee: avcodec/dpx: Move need_align to act per line [12:58:08 CET] ffmpeg 03Michael Niedermayer 07release/2.4:3879202d68ce: avcodec/error_resilience: avoid accessing previous or next frames tables beyond height [12:58:09 CET] ffmpeg 03Michael Niedermayer 07release/2.4:0814b140b1d7: avcodec/dxtory: Fix input size check in dxtory_decode_v1_420() [12:58:10 CET] ffmpeg 03Michael Niedermayer 07release/2.4:6ba69f60ecb6: avcodec/dxtory: Fix input size check in dxtory_decode_v1_410() [12:58:11 CET] ffmpeg 03Michael Niedermayer 07release/2.4:9356635e761a: avcodec/takdec: Skip last p2 sample (which is unused) [12:58:12 CET] ffmpeg 03Michael Niedermayer 07release/2.4:3449b47dc548: avcodec/smacker: Check that the data size is a multiple of a sample vector [12:58:13 CET] ffmpeg 03Michael Niedermayer 07release/2.4:dccb80dd50ee: avcodec/wmaprodec: Check for overread in decode_packet() [12:58:14 CET] ffmpeg 03Michael Niedermayer 07release/2.4:aa780a52717e: avcodec/jpeg2000: Use av_image_check_size() in ff_jpeg2000_init_component() [12:58:15 CET] ffmpeg 03Michael Niedermayer 07release/2.4:9aa4b9c2a342: avcodec/jpeg2000: Check comp coords to be within the supported size [12:58:16 CET] ffmpeg 03Michael Niedermayer 07release/2.4:50870dd3de5c: avcodec/jpeg2000dec: Check SIZ dimensions to be within the supported range [12:58:17 CET] ffmpeg 03Michael Niedermayer 07release/2.4:f1058efc8160: avcodec/jpeg2000dec: Fix potential integer overflow with tile dimensions [12:58:18 CET] ffmpeg 03Michael Niedermayer 07release/2.4:3cef69c576c5: avformat/utils: Do not init parser if probing is unfinished [12:58:19 CET] ffmpeg 03Michael Niedermayer 07release/2.4:75fa9c0b39f1: avformat/matroskadec: Check subtitle stream before dereferencing [12:58:20 CET] ffmpeg 03Martin Storsj� 07release/2.4:65d6748d9e42: rtmpcrypt: Do the xtea decryption in little endian mode [12:58:21 CET] ffmpeg 03Michael Niedermayer 07release/2.4:ee5ba259d12d: avcodec/vp8: Do not use num_coeff_partitions in thread/buffer setup [12:58:22 CET] ffmpeg 03Michael Niedermayer 07release/2.4:e9ec9be14679: avcodec/h264_slice: Limit max_contexts when slice_context_count is initialized [12:58:23 CET] ffmpeg 03Michael Niedermayer 07release/2.4:e8054a9595fe: avcodec/cabac_functions: Fix "left shift of negative value -31767" [12:58:24 CET] ffmpeg 03Michael Niedermayer 07release/2.4:20de3b007bbd: avcodec/cabac: Check initial cabac decoder state [12:58:25 CET] ffmpeg 03Michael Niedermayer 07release/2.4:961a1f73e392: avcodec/hevc: Check entry_point_offsets [12:58:26 CET] ffmpeg 03Michael Niedermayer 07release/2.4:c44ee37cbd05: avcodec/jpeg2000dwt: Check ndeclevels before calling dwt_decode*() [12:58:27 CET] ffmpeg 03Michael Niedermayer 07release/2.4:b6932f643438: avcodec/jpeg2000dwt: Check ndeclevels before calling dwt_encode*() [12:58:28 CET] ffmpeg 03Michael Niedermayer 07release/2.4:453542f14d3c: avcodec/hevc_cabac: Fix multiple integer overflows [12:58:29 CET] ffmpeg 03Michael Niedermayer 07release/2.4:02764f12a671: avcodec/hevc: allocate entries unconditionally [12:58:30 CET] ffmpeg 03Michael Niedermayer 07release/2.4:af3e5bdd0d32: avcodec/vp3: Clear context on reinitialization failure [12:58:31 CET] ffmpeg 03Michael Niedermayer 07release/2.4:733510fb6583: avcodec/utils: Use 64bit for aspect ratio calculation in avcodec_string() [12:58:32 CET] ffmpeg 03Michael Niedermayer 07release/2.4:95bdbfe85159: avcodec/utils: Clear dimensions in ff_get_buffer() on failure [12:58:33 CET] ffmpeg 03Michael Niedermayer 07release/2.4:a3ec4b307b4f: avformat/dump: Fix integer overflow in av_dump_format() [12:58:34 CET] ffmpeg 03Michael Niedermayer 07release/2.4:5cc411022e49: avutil/integer: Fix av_mod_i() with negative dividend [12:58:35 CET] ffmpeg 03Michael Niedermayer 07release/2.4:53ddc450c80f: avutil/mathematics: Do not treat INT64_MIN as positive in av_rescale_rnd [12:58:36 CET] ffmpeg 03Michael Niedermayer 07release/2.4:c3f268b01c89: avcodec/mpeg4videodec: Check available data before reading custom matrix [12:58:37 CET] ffmpeg 03Michael Niedermayer 07release/2.4:516525a1031b: avcodec/vp3: always set pix_fmt in theora_decode_header() [12:58:38 CET] ffmpeg 03Michael Niedermayer 07release/2.4:ce15d773d4f7: avcodec/apedec: Check length in long_filter_high_3800() [12:58:39 CET] ffmpeg 03Michael Niedermayer 07release/2.4:f91e71cc68fb: avutil/mathematics: return INT64_MIN (=AV_NOPTS_VALUE) from av_rescale_rnd() for overflows [12:58:40 CET] ffmpeg 03Michael Niedermayer 07release/2.4:4ecdd45d2410: avutil/timecode: Fix fps check [12:58:41 CET] ffmpeg 03Timo Ter�s 07release/2.4:ec94195b80d9: mpegencts: Fix overflow in cbr mode period calculations [12:58:42 CET] ffmpeg 03Michael Niedermayer 07release/2.4:1aa415ffb152: avcodec/vp3: Fix "runtime error: left shift of negative value" [12:58:43 CET] ffmpeg 03Michael Niedermayer 07release/2.4:0819598b008e: avformat/smacker: fix integer overflow with pts_inc [12:58:44 CET] ffmpeg 03Michael Niedermayer 07release/2.4:f19d3fe8e951: avcodec/wmaprodec: Fix overflow of cutoff [12:58:45 CET] ffmpeg 03Michael Niedermayer 07release/2.4:ae434647409d: avcodec/wmaprodec: Check bits per sample to be within the range not causing integer overflows [12:58:46 CET] ffmpeg 03Michael Niedermayer 07release/2.4:142f8308a121: avcodec/dirac_parser: Fix potential overflows in pointer checks [12:58:47 CET] ffmpeg 03Michael Niedermayer 07release/2.4:e5a2128ead24: avcodec/dirac_parser: Add basic validity checks for next_pu_offset and prev_pu_offset [12:58:48 CET] ffmpeg 03Michael Niedermayer 07release/2.4:9459490c2ae9: avcodec/dirac_parser: Check that there is a previous PU before accessing it [12:58:49 CET] ffmpeg 03Michael Niedermayer 07release/2.4:bafd5c3c80a4: avcodec/hevc: Fix integer overflow of entry_point_offset [12:58:50 CET] ffmpeg 03Michael Niedermayer 07release/2.4:c02a9f1c6e8c: swscale/utils: Fix for runtime error: left shift of negative value -1 [12:58:51 CET] ffmpeg 03Michael Niedermayer 07release/2.4:c676db730e36: avcodec/pgssubdec: Fix left shift of 255 by 24 places cannot be represented in type int [12:58:52 CET] ffmpeg 03Michael Niedermayer 07release/2.4:311de799a81e: avcodec/jpeg2000dec: Check bpno in decode_cblk() [12:58:53 CET] ffmpeg 03Michael Niedermayer 07release/2.4:9ee155c745b5: avcodec/vp3: ensure header is parsed successfully before tables [12:58:54 CET] ffmpeg 03Michael Niedermayer 07release/2.4:70720f3e4d9e: avcodec/hevc: Check max ctb addresses for WPP [12:58:55 CET] ffmpeg 03Michael Niedermayer 07release/2.4:6c83283c3a00: avcodec/ffv1dec: Print an error if the quant table count is invalid [12:58:56 CET] ffmpeg 03Michael Niedermayer 07release/2.4:3a1aaec9bbcc: avcodec/ffv1dec: Clear quant_table_count if its invalid [13:46:01 CET] ffmpeg 03Michael Niedermayer 07release/2.4:9f0e36b10116: update for 2.4.12 [13:52:20 CET] atomnuker, did you see "http://fate.ffmpeg.org/report.cgi?time=20151206103056&slot=x86_32-debian-kfreebsd-gcc-4.1" <-- aac-pns-encode failure [13:53:24 CET] same failure also occurs with more recent gccs like gcc 4.4.5 (Debian 4.4.5-8) [13:54:19 CET] also gcc 4.9.3 (Gentoo 4.9.3 p1.2, pie-0.6.3) [13:54:30 CET] http://fate.ffmpeg.org/report.cgi?time=20151206110425&slot=x86_32-athlon64-gentoo-gcc-3dnow [13:57:24 CET] michaelni: at first I thought they were miscompiling but now I think about it I think the reference needs to be changed [14:04:23 CET] isnt the test encodig a wav then decoding aand comparing to that wav ? or am i mixing something up? [14:06:09 CET] michaelni: the reference definitely changed, that's why they fail [14:06:30 CET] you mean the .wav changed ? [14:06:59 CET] no, CMP_TARGET did [14:09:05 CET] ahh yes, in ca203e9985cd2dcf42a0c0853940850d3a8edf3a [14:10:13 CET] i guess ill leave that to you and claudio to fix [14:11:05 CET] btw, if i should contact claudio about it, tell me [14:12:14 CET] michaelni: no, there was a commit later that broke it [14:12:37 CET] I remember FATE said it last succeeded before a Ganesh commit which improved the table accuracy [14:13:12 CET] and since the deviation is failing because it's lower it would make sense [14:14:01 CET] either way I'm getting CMP value of 616 on my machine rather than the 663 as is currently in the test [14:14:25 CET] I'll give it a test on another machine and I'll push a fix for that [14:14:50 CET] (along with some TNS fixes which solve the problems in the recent trac ticket) [14:16:03 CET] ffmpeg 03Michael Niedermayer 07release/2.0:58986f0fa151: avcodec/ffv1dec: Check for 0 quant tables [14:16:04 CET] ffmpeg 03Michael Niedermayer 07release/2.0:5628d6202217: avcodec/ffv1dec: Print an error if the quant table count is invalid [14:16:05 CET] ffmpeg 03Michael Niedermayer 07release/2.0:2789b48b4ef5: avcodec/ffv1dec: Clear quant_table_count if its invalid [14:16:07 CET] ffmpeg 03Michael Niedermayer 07release/2.1:fde9b80c17e1: avcodec/ffv1dec: Check for 0 quant tables [14:16:08 CET] ffmpeg 03Michael Niedermayer 07release/2.1:d9593e075161: avcodec/ffv1dec: Print an error if the quant table count is invalid [14:16:09 CET] ffmpeg 03Michael Niedermayer 07release/2.1:65f4425ad580: avcodec/ffv1dec: Clear quant_table_count if its invalid [14:16:10 CET] ffmpeg 03Michael Niedermayer 07release/2.2:1aa1f4bbf45e: avcodec/ffv1dec: Check for 0 quant tables [14:16:11 CET] ffmpeg 03Michael Niedermayer 07release/2.2:8a6a619cbeee: avcodec/ffv1dec: Print an error if the quant table count is invalid [14:16:12 CET] ffmpeg 03Michael Niedermayer 07release/2.2:a9e4be1cdf6f: avcodec/ffv1dec: Clear quant_table_count if its invalid [14:16:13 CET] ffmpeg 03Michael Niedermayer 07release/2.3:eb0a9cd68a25: avcodec/ffv1dec: Check for 0 quant tables [14:16:14 CET] ffmpeg 03Michael Niedermayer 07release/2.3:1fff9947098b: avcodec/ffv1dec: Print an error if the quant table count is invalid [14:16:15 CET] ffmpeg 03Michael Niedermayer 07release/2.3:0fde898f16cf: avcodec/ffv1dec: Clear quant_table_count if its invalid [14:22:06 CET] ffmpeg 03Ganesh Ajjanagadde 07master:90409b6da88b: avformat/rtsp: free opts dictionary on failure of getnameinfo [14:22:07 CET] ffmpeg 03Ganesh Ajjanagadde 07master:f1f323b6de03: avformat/movenc-test: correct varargs usage [14:22:08 CET] ffmpeg 03Ganesh Ajjanagadde 07master:14886bebfe7a: avcodec/dvdsubdec: fix typo in dlog message [14:22:09 CET] ffmpeg 03Ganesh Ajjanagadde 07master:65440916c17d: ffplay: use hypot() [14:26:53 CET] alright, some (but not all) of my instances are going back. [14:54:27 CET] wow even gstreamer dropped Libav [14:56:28 CET] wow [14:58:26 CET] kierank: I don't seem to have access to avdev (permission denied (publickey)) [14:58:33 CET] so I can't test my patches there [14:58:43 CET] yeah I reinstalled the machine [14:59:46 CET] I've uploaded them to the ML, so if people can see if the retuned CMP_TARGETs make FATE pass on the kfreebsd machines that'd be nice [15:00:50 CET] atomnuker: I'll try and setup the vpn tomorrow [15:00:50 CET] so you can use internaldev [15:02:01 CET] michaelni: ping since you test a system with a failing PNS on FATE (x86_32-debian-kfreebsd-gcc-4.4) [15:03:41 CET] ffmpeg 03Timothy Gu 07master:9670165cc227: ffprobe: Do not print profile names in -bitexact [15:03:42 CET] ffmpeg 03Timothy Gu 07master:b0669e7ccfef: fate/concatdec: Use -bitexact [15:09:47 CET] coverage.ffmpeg.org in sync again [15:10:14 CET] btw, small reminder, for anyone curious about the evolution of the code coverage, there is http://lucy.pkh.me/ffmpeg-coverage-snapshots/ [15:13:04 CET] mateo`: http://fate.ffmpeg.org/report.cgi?time=20151206141037&slot=x86_64-archlinux-gcc-asan [15:13:16 CET] atomnuker, i dont see it on the ML yet but probably easier to push and see if fate gets fixed then trying to reproduce the issue outside a fate client [15:13:29 CET] its not happening locally on x86-32 it seems [15:14:28 CET] atomnuker, seems its reproduceable with --disable-asm on x86-32 [15:14:36 CET] ill test your patch once it appears [15:14:42 CET] mateo`: also, http://fate.ffmpeg.org/report.cgi?time=20151206135755&slot=x86_64-archlinux-gcc-disablezlib [15:14:50 CET] it seems to be missing a dependency check [15:15:15 CET] atomnuker, whats the subj of the patch/mail btw? [15:15:35 CET] aacenc: move the TNS search and filtering before PNS [15:15:47 CET] the other 2 patches are replies to this one [15:17:26 CET] am i the only one who hasnt received these mails yet ? [15:17:38 CET] they are in the ML acrhive [15:17:56 CET] mails are taking some time to reach me too [15:18:57 CET] TimNich, theres a mail latency problem (see above) (not sure this is a new or still a existing one) but it definitly needs to be fixed [15:20:31 CET] i didnt get the mails yet either [15:20:52 CET] TimNich, is there something i can do or look into to help about this mail issue ? [15:21:58 CET] just got a mail datestamped 50 minutes ago [15:22:00 CET] http://fate.ffmpeg.org/report.cgi?time=20151206142043&slot=x86_64-archlinux-gcc-ubsan erh.. [15:22:00 CET] something surely is slow [15:23:02 CET] would this be because of "type const" vs "const type"? [15:23:26 CET] int32_t const *samples_s32[2] = { (const int32_t *)samples0, [15:23:28 CET] (const int32_t *)samples1 }; [15:23:42 CET] maybe we should write const int32_t *samples_s32? [15:24:16 CET] const type *var and type const *var are semantically identical [15:24:27 CET] yes, but maybe ubsan is confused? [15:25:18 CET] i personally prefer the const before the type myself [15:25:37 CET] i think that's what we use all over the code base anyway [15:25:57 CET] not sure why index checks should fail from that, however [15:29:31 CET] can I be added to ffmpeg-security btw [16:13:33 CET] TimNich, where does the spamassasin logs go ? [16:30:07 CET] nevcairiel: vp9 tests are failing with threading [16:30:26 CET] (not sure if it's because of your recent commits but i think it was related) [16:30:34 CET] http://fate.ffmpeg.org/report.cgi?time=20151206144611&slot=x86_64-archlinux-gcc-threads-8 [16:30:37 CET] what did i do to vp9 [16:30:55 CET] didn't you add accel or sth? [16:31:00 CET] not yet [16:31:03 CET] ah [16:31:10 CET] failure is probably related to bbb's recent changes [16:31:12 CET] then maybe i should pokeBBB [16:31:16 CET] yeah [16:31:27 CET] it was time i needed to renable my fate instances... [16:31:35 CET] so much issues :p [16:31:45 CET] and valgrind instance not yet done [16:33:01 CET] vp9_decode_update_thread_context probably needs to check for pix_fmt changes [16:35:16 CET] more specifically, just copy s->pix_fmt between threads [16:38:42 CET] lets test if i can fix it and send a patch [16:43:16 CET] nice the new hetzner offer https://www.hetzner.de/ot/hosting/produkte_rootserver/ex41 [16:43:48 CET] wow [16:44:04 CET] those things are getting cheaper every year [16:44:06 CET] that's pretty nice as far as dedis go [16:44:15 CET] i have a haswell variant from 2013 or so [16:44:39 CET] I will see when I'll want to go over the 20eur I pay right now for my 8core atom [16:44:55 CET] skylake 2x4TB at this price is pretty nice [16:45:00 CET] btw, i see some ARM boxes as well [16:45:15 CET] maybe i should order one for real arm fate instances [16:45:32 CET] ubitux: i send a patch for vp9 to the ml [16:45:42 CET] ubitux, dont we have lots of arm instance [16:46:15 CET] Daemon404: qemu mostly [16:46:21 CET] nevcairiel: can't really review [16:48:32 CET] ubitux, every one i looked at just now seems real [16:49:53 CET] ubitux: didnt you write half the decoder :D [16:50:02 CET] not really [16:50:20 CET] that's just what Ronald like to say :) [16:54:35 CET] wow SMC. [16:57:46 CET] I wonder if I should send a patch for nvenc that removes the nonfree condition but instead requires the most recent version [16:57:53 CET] making nonfree version dependent seems rather complex [16:57:59 CET] huh? [16:58:08 CET] Did they actualy fix their shit? [16:58:10 CET] they released the 6.0 sdk [16:58:15 CET] the header is now MIT licensed [16:58:19 CET] wow [16:58:27 CET] so it could even be bunbled, removing the dependency entirely? [16:58:30 CET] -b [16:58:42 CET] license wise, sure, i guess [16:58:59 CET] I guess i'll just do that and send it to the ML [16:59:27 CET] could just be called nvenc.h [17:00:16 CET] woah, and the SDK is only 20MB now, no longer a 600MB blob, like the SDK v4 [17:00:28 CET] v5 already cleaned out a lot of the stuff [17:00:51 CET] they also added the cuvid decoder into this sdk, so you dont need the full cuda sdk blob anymore [17:01:08 CET] Why would anyone still use that? [17:01:15 CET] dunno [17:01:36 CET] The download is slow as hell though [17:01:38 CET] the API is extremely simple to use, so for a quick nvidia transcode app it might work [17:02:06 CET] Currently at an amazing 50KB/s [17:02:17 CET] didnt notice slowness when i grabbed it earlier [17:02:41 CET] down to 30 now [17:03:15 CET] just tried again, download still fast [17:03:21 CET] the hell [17:03:58 CET] looks like only one of their servers is dead [17:09:20 CET] So does that mean nvenc would now be enabled by default in lgpl builds? [17:09:25 CET] So basicaly, removing all checks? [17:09:46 CET] if it doesnt add runtime dependencies, then sure [17:09:57 CET] All runtime deps are dlopene'd [17:10:08 CET] but i'm sure someone will complain about bundling the nvidia header somewhere =p [17:10:56 CET] From how I understand the MIT license, I'm free to do whatever I want with it. So if i'd make a stripped down version of it that complies with the ffmpeg style, why not? [17:11:28 CET] personally I would probably use the untouched header, makes updating it in the future easier [17:11:51 CET] I wouldn't change how it works, just style it propperly and remove most comments [17:12:01 CET] And add Cygwin support ;) [17:12:37 CET] people still care about cygwin? [17:12:49 CET] It's my primary "platform", so I do. [17:13:32 CET] All that's missing is a single check, that sets the stdcall convention when building on cygwin [17:13:50 CET] does cygwin not set _WIN32? [17:14:01 CET] of course it wouldnt [17:14:06 CET] its supposed to emulate linux [17:15:03 CET] This configure script... I hate it every time i have to touch it. [17:15:23 CET] its not that bad when you get to know where things are [17:15:39 CET] Currently searching what I have to change to enable nvenc by default [17:16:04 CET] just remove the external nvenc component entirely [17:16:24 CET] and treat it like an internal decoder [17:16:27 CET] encoder* [17:16:44 CET] It still needs ldl though [17:17:19 CET] https://gist.github.com/BtbN/18ffe94bde0d84286c52 so, like this? [17:18:38 CET] enabling it without any sort of tests might be risky though, who knows if the header works on every system out there, and there certainly arent binaries for every system [17:20:08 CET] So basicaly just a "compile #include "nvenc.h"" test? [17:20:32 CET] And limited to windows and linux [17:22:55 CET] you could keep the generic nvenc component so it can be disabled entirely with the checks, and just enable it by default (add it somewhere around line 2980, maybe to the xlib line, its so lonely) [17:23:26 CET] wait wee still use libx11/xlib? [17:23:35 CET] i thought we moved to xcb based off of vlc's [17:23:35 CET] of course [17:23:51 CET] x11grab uses xlib [17:24:01 CET] You mean, leaving in nvenc_encoder_deps="nvenc", call enable nvenc somewhere, and then do checks that disable it? [17:24:02 CET] we didnt backport xcbgrab? [17:24:04 CET] :V [17:24:06 CET] and the vaapi ffmpeg.c module uses xlib [17:24:16 CET] There is a vaapi ffmpeg.c module? [17:24:43 CET] hm there isnt, whats this check used for then [17:24:57 CET] propably vdpau, as it also needs an X display? [17:25:20 CET] there is HAVE_VAAPI_X11, but its not used anywhere, apparently [17:25:24 CET] vaapi doesn't need xlib [17:25:33 CET] Depends on how you use it [17:25:38 CET] For ffmpeg, it doesn't. [17:25:39 CET] you can also use it with a wayland display, or use raw drm [17:26:25 CET] a ffmpeg_vaapi.c (if it ever exists) should probably use raw drm [17:27:31 CET] Daemon404: we have xcbgrab, but what makes you think that replaced x11grab? =p [17:28:02 CET] nothing ever gets removed etc. [17:28:17 CET] nevcairiel, ... because it is better in every way? [17:28:20 CET] and just as portable [17:28:28 CET] what do i care about files that never get compiled for me, if you care, send a removal patch =p [17:28:29 CET] NOBODY is going to be missing libxcb [17:29:17 CET] what if someone has a SunOS machine from 1986? [17:32:29 CET] Hm, there is no appropriate check for a header that's part of ffmpeg [17:32:48 CET] no, that concept is slightly odd =p [17:47:20 CET] I'm not even sure how to propperly include that header. Is it safe to assume that the conftest scripts are built in the toplevel source dir? [17:47:47 CET] no [17:47:54 CET] out off tree builds are normal [17:48:01 CET] there's a var for that [18:51:21 CET] TimNich, are you around ? ive figured out how to make mailman be verbose about SMTP connections but you know the mail stuff i dont [19:08:38 CET] configure on cygwin takes forever... [19:09:16 CET] even longer than on windows? =p [19:09:44 CET] it forks _a lot_ [19:09:48 CET] and fork on cygwin is slow [19:10:19 CET] msys doesn't try to emulate fork like cygwin does, so it's quite a bit faster. [19:10:57 CET] ffmpeg 03Paul B Mahol 07master:45b451c89239: avfilter/af_alimiter: add 3 more options [19:11:19 CET] mingw is already quite slow, if you compare to linux.. [19:11:50 CET] looks like nobody tried to build on cygwin for a while. The schannel stuff explodes. It propably shouldn't even try to build it. [19:12:36 CET] i think i saw someone post about that on trac [19:12:46 CET] but cba to cygwin [19:12:57 CET] it spews so many errors that it exceeds my 50k line buffer in my terminal. [19:13:43 CET] why do you need to ue cygwin anyway? [19:13:50 CET] Because i want to use it. [19:14:00 CET] And as it's a supported platform, stuff should build. [19:14:31 CET] its a "supported" platform, one of those no core dev cares about or tests on =p [19:14:54 CET] and its crazy enough to break easily [19:15:28 CET] i don't see the appeal when you can build natively for windows, and even with pthreads emulation if that's what you want, but to each their own [19:15:49 CET] I use Cygwin to have a propper terminal on Windows [19:16:12 CET] you can use the cygwin environment and build using mingw [19:16:26 CET] they have a mingw-w64 package [19:16:26 CET] Sure, but then ffmpeg isn't aware of cygwin-paths [19:18:24 CET] BtbN: https://trac.ffmpeg.org/ticket/5036 [19:18:25 CET] huh, aparently cygwin doesn't emulate fork anymore [19:18:38 CET] they use the actual fork function from some low-level kernel interface [19:19:39 CET] Was that patch ever sent? [19:20:49 CET] apparently not [19:20:53 CET] BtbN: wow, do they? reference for this? [19:21:29 CET] BtbN: probably not [19:21:47 CET] if it works, should send it to the ML [19:35:55 CET] TimNich, i maybe understand what is causing the latency issues, but not sure, i might be wrong too [20:33:58 CET] someone's really passionate about Apple Graphics [20:38:38 CET] he is just too stubborn to simply post the info requested =p [20:39:09 CET] i think its funny that users can keep re-opening shit a developer closed, though [20:46:43 CET] looks like he remuxed http://samples.mplayerhq.hu/V-codecs/SMC/aletrek-smc.mov with mkvmerge [20:46:51 CET] and that remuxed file doesn't work [21:03:33 CET] Hi! Does git clone http://git.videolan work for anybody? [21:03:40 CET] What was the reason for the change? [21:04:31 CET] from http:// to git:// you mean ? [21:04:42 CET] have to ask in #videolan probably [21:04:52 CET] or maybe j-b.. [21:06:16 CET] But I am curious if it works for FFmpeg: After all, we changed our homepage and since I didn't see any discussion on ffmpeg-devel, I assumed the discussion was here, wasn't it? [21:07:58 CET] works for me, but cloning through http is not as verbose, so you wont get the typical status reporting as you used to get ... so just need to trust it works [21:08:32 CET] Well, it is verbose here, I get a lot of error reports (after waiting for a very long time) [21:08:41 CET] eror messages. [21:09:00 CET] should talk to thresh then [21:16:59 CET] ffmpeg 03Rostislav Pehlivanov 07master:b32e989e6c65: aacenc: move the TNS search and filtering before PNS [21:17:00 CET] ffmpeg 03Rostislav Pehlivanov 07master:d55f83de4d41: aacenc_tns: tune and reduce artifacts [21:17:01 CET] ffmpeg 03Rostislav Pehlivanov 07master:a0a39acd01ea: fate: change the CMP_TARGETs for the recent AAC encoder changes [21:17:21 CET] RiCON: I know that you want to help but how does the command line you provided allow to reproduce the issue in question? [21:17:35 CET] For the attached file [21:18:17 CET] the issue's name comes from a comment tag in the matroska file [21:18:39 CET] i don't have ffplay compiled, compiling now so i can test with it [21:19:30 CET] Please do not use ffplay! [21:19:38 CET] Just use "ffmpeg -i input -qscale 2 out.avi" [21:19:55 CET] oh ok, you want the resulting file [21:20:17 CET] FFplay depends on an external library that sometimes is responsible for FFplay issues. [21:20:21 CET] no, but you can test that way and play the avi [21:20:52 CET] Hey, Carl's back on irc! [21:21:28 CET] ffmpeg doesn't really complain encoding or playing the .avi though [21:21:43 CET] it's just black instead of showing what's in the original file [21:22:14 CET] That's fine. [21:22:31 CET] I mean this is what I need to understand the ticket. [21:23:29 CET] thresh does not react: May I revert the homepage patch? [21:23:35 CET] Hi, atomnuker! [21:24:00 CET] Or can somebody explain the reasoning, I probably miss something...? [21:24:20 CET] the reasoning is that videolan would prefer if people used http instead [21:24:20 CET] cehoyos: you mean the git:// vs http:// one? [21:25:03 CET] yeah, that's pretty much it, thresh said he had bad memories of git:// and wanted to minimize usage on the videolan network [21:25:29 CET] This is exactly my question: What difference does it make? [21:25:51 CET] git-daemon is shitty is the difference, and they dont want so much users on it [21:25:51 CET] Apart from the fact that I do not remember an issue with git:// (reported by users or myself) and http:// just doesn't work here. [21:26:23 CET] git:// is faster [21:26:41 CET] Is it really so difficult to understand what I request? [21:26:43 CET] though the main limit is still bandwidth [21:27:06 CET] RiCON: I never copy and paste but still: If you are a native speaker, please suggest a better wording... [21:28:12 CET] atomnuker: no, it should not. [21:29:15 CET] i'm not native speaker, but if you're referring to the original person who opened the issue, i have no idea what was required to work [21:30:19 CET] What I mean is: I did not request attaching an output file (the issue is that this usually leads to developers testing the output file and not understanding the issue), I simply tried to ask for a command line that allows to reproduce the issue pasted together with its console output. [21:31:38 CET] j-b: git clone fails here with the new url, anything that can be done about it? [21:32:25 CET] ok, so just a `mpv -v aletrek.mkv` followed by console output would be enough? [21:32:46 CET] cehoyos: ask thresh, he's around. [21:32:56 CET] cehoyos: the old urls still work. [21:34:30 CET] I opened a channel with him, he did not react so far. [21:34:44 CET] The problem is that the new url (that does not work here) is on the FFmpeg homepage. [21:35:03 CET] j-b: how come? [21:35:08 CET] (I was unable to find an explanation or discussion for the homepage change, only that it does not work). [21:35:27 CET] RiCON: No, because this isn't mpv-bugs... [21:35:49 CET] (And because - iirc - I asked for an "FFmpeg command line that allows ...") [21:36:16 CET] cehoyos: was discussed yesterday [21:36:20 CET] j-b: definitely seems very slow right now to clone over http [21:36:21 CET] atomnuker: http is cachable [21:36:25 CET] On irc [21:36:28 CET] nevcairiel: ask thresh [21:38:39 CET] Anyone looking for paid asm work BTW? [21:40:46 CET] May I suggest to revert until the issues are solved? [21:41:02 CET] Or is there a big issue with git:// that I do not understand? [21:41:42 CET] I can revert when I get home yes [21:51:06 CET] Please do. [21:54:00 CET] atomnuker: Year-long experience has tought us to close a ticket and request to reopen if it is not fixed... [21:59:19 CET] right, mpv still doesn't work with the mkv directly because it uses its own matroska demuxer. [21:59:33 CET] forcing lavf as demuxer it plays the sample file fine [22:27:03 CET] cehoyos: reverted [22:33:48 CET] kierank: what kind of asm? [22:34:14 CET] some x262 and some dirac [22:34:31 CET] for x86? [22:34:40 CET] yes [22:34:51 CET] only dirac is within ffmpeg? [22:35:06 CET] yes [22:35:11 CET] i think durandal_1707 was looking for some asm to write recently [22:36:08 CET] if its simple enough [22:36:09 CET] i see some existing asm for dirac, what's missing? [22:36:22 CET] 10-bit [22:36:49 CET] and dequantisation [22:37:05 CET] note: i won't have time, i'm just curious [22:37:37 CET] But the 10-bit code is missing from c as well, or am I wrong? [22:37:58 CET] cehoyos: see my latest patchset, more will be coming [22:42:53 CET] ffmpeg 03Andreas Cadhalpun 07master:5adb5d9d894a: mjpegdec: consider chroma subsampling in size check [23:04:46 CET] Re: cygwin. I haven't had my desktop turned on early enough to catch the old fate job I had scheduled. [23:04:50 CET] BtbN ^ [23:06:12 CET] I really should have changed the time. [23:57:18 CET] kierank: so how is cfhd going? [23:59:45 CET] durandal_1707: kinda ok, a bit confused about how to filter out junk bytes on new samples [23:59:56 CET] the junk bytes could clash with valid tags [00:00:00 CET] --- Mon Dec 7 2015 From burek021 at gmail.com Mon Dec 7 02:05:01 2015 From: burek021 at gmail.com (burek) Date: Mon, 7 Dec 2015 02:05:01 +0100 (CET) Subject: [Ffmpeg-devel-irc] ffmpeg.log.20151206 Message-ID: <20151207010502.6CBB818A00B8@apolo.teamnet.rs> [00:00:45 CET] or at least thats what I think I read/was told [00:01:09 CET] I think the issue I'm running into is I'm trying to scale down as well as pad, in the same command... [00:02:27 CET] satt -> well man, im reading the docs about pad and it is as i told you pad=640:480:0:40:violet Add paddings with the color "violet" to the input video. The output video size is 640x480, and the top-left corner of the input video is placed at column 0, row 40 [00:02:41 CET] that is w:h:x:y structure isnt it? [00:04:03 CET] if you put 0:0:320:568 as input area, you have a 0x0 [00:04:34 CET] i have croped and scale after with no problem [00:07:19 CET] totally is yeah. what I'm saying is that since x/y default to 0 you don't need to include them. my command works for most videos, except for when the input file is larger than the desired output like the error message says [00:07:39 CET] Input area 0:0:320:568 not within the padded area 0:0:320:480 or zero-sized [00:08:32 CET] 0:0:320:568 would be padding to 0x0, positioning the video at 320x568 - at least from what I understand [00:09:46 CET] I appreciate the help man! but gotta run for a bit, back later [00:10:57 CET] what help :/ jajaja [00:11:00 CET] have a great day [01:11:14 CET] back. hahah I appreciate you trying to help at least [01:28:02 CET] Anyone know how I can solve this issue? I think it's caused from an input being larger than the area I'm padding to. I'm also scaling. Here's the error [01:28:10 CET] Input area 0:0:320:568 not within the padded area 0:0:320:480 or zero-sized [01:28:21 CET] command: ffmpeg -i INPUTS -filter_complex [0:v]scale=320:-2,pad=320:480,setsar=sar=1/1[v0]; [1:v]scale=320:-2,pad=320:480,setsar=sar=1/1[v1]; [2:v]scale=320:-2,pad=320:480,setsar=sar=1/1[v2]; [3:v]scale=320:-2,pad=320:480,setsar=sar=1/1[v3]; [4:v]scale=320:-2,pad=320:480,setsar=sar=1/1[v4]; [v0] [0:a:0] [v1] [1:a:0] [v2] [2:a:0] [v3] [3:a:0] [v4] [4:a:0] concat=n=5:v=1:a=1 [v] [a] -map [v] -map [a] out.mp4 [01:34:04 CET] scale=if(gt(ih, iw), -2, 320):if(gt(ih, iw), 480, -2) [01:34:11 CET] instead of your current scale filters [01:34:13 CET] (probably) [01:34:18 CET] Note, you may need to escape things [01:45:03 CET] ah thanks c_14, I'll give that a shot. Should I be keeping the pad arguments? [01:45:11 CET] yes [01:45:36 CET] ok sweet [01:58:11 CET] c_14: Looks like I need to escape or something, here's what I'm getting [01:58:19 CET] Invalid size 'if(gt(ih' [01:58:33 CET] escape the comma [01:58:54 CET] with a \ ? [01:59:38 CET] yes [02:01:29 CET] cool thanks. it's running now [02:05:49 CET] c_14 thanks so much! seems to have solved it!! [04:20:46 CET] hello, need help in compiling ffmpeg on debian 8 [04:21:13 CET] im using this guide https://trac.ffmpeg.org/wiki/CompilationGuide/Ubuntu but towards the end i am getting this error "ERROR: x265 not found using pkg-config" [04:21:17 CET] any suggestions? [04:21:59 CET] luke11: https://bitbucket.org/multicoreware/x265/issues/212/new-libx265-not-found-in-ffmpeg-configure [04:22:46 CET] thanks will give it a shot [04:33:34 CET] relaxed, didn't work for me unfortunately [04:34:54 CET] not sure where exactly i need to add in the x265.h file [04:36:35 CET] change it to --> #include [04:36:43 CET] will try, ty [04:43:44 CET] hey everyone.. i am using ffmpeg hls segmenting with AES encryption... i have the whole ffmpeg stream with while 1 do ffmpeg ..... etc etc.. so basically if the stream fails it will reinitialize... the problem is sometimes i can't decrypt the channels and i have to restart them again for the decryption to happen again.. anyone know why its doing that.. if ffmpeg is restarted in the same session with the same script and stuff should it not [04:43:44 CET] maintain the same keyinfo file ? [04:49:05 CET] Admin_234234: hallo [04:49:14 CET] hi [04:49:21 CET] Admin_234234: wie geht's es dir ? [04:49:31 CET] ? [05:55:14 CET] Does ffmpeg support 3gp & 3gp2? - I'm trying to find a complete listing of supported containers. [06:01:22 CET] aphorise: it should. check ffmpeg -formats [06:09:31 CET] c_14 thanks - do you think all video formats report a duration or is that attainable with: ffprobe -i sample.webm -select_streams v -show_streams 2>&1 | grep Duration # ? [16:21:39 CET] `ffmpeg -formats` which video formats would potentially not provide its Durations? - I just trying to finalise a wrapper script using ffprobe && ffmpeg to confirm validity of files using -1 seconds less than the claimed length - just wondering which sorts of general `*.extension` I can enable as guide to supported container formats like `*.avi` which are supported for Demuxing [16:40:02 CET] hey [19:23:19 CET] omg there are so many potential file extension that ffmpeg could parse ... I was going through a few of my old files & old line listings [19:35:05 CET] Actually the vlc listing for videos seems to be the most sane + a few more that I have on my list [19:38:15 CET] hey everyone.. check this out " http://pastebin.com/f1qHT061 " .. so i have the ffmpeg portion in a loop.. i am using the encryption and files exist.. why is it sometimes when ffmpeg restarts using the while 1 loop the encryption stops working? i hvae to then force a restart again to get the image back. It's as if when it restarts the ffmpeg using the while 1 loop, it breaks and the system doesnt know to encrypt using the key .. weird.. [19:38:15 CET] doesn't exist complaining tha thte keyinfo file and key file doesnt' exist.. very odd.. any suggestions ? [19:40:00 CET] hey everyone.. check this out " http://pastebin.com/f1qHT061 " .. so i have the ffmpeg portion in a loop.. i am using the encryption and files exist.. why is it sometimes when ffmpeg restarts using the while 1 loop the encryption stops working? i hvae to then force a restart again to get the image back. It's as if when it restarts the ffmpeg using the while 1 loop, it breaks and the system doesnt know to encrypt using the key .. [19:40:01 CET] weird..doesn't exist complaining tha thte keyinfo file and key file doesnt' exist.. very odd.. any suggestions ? [20:00:36 CET] anyone [20:00:37 CET] ? [20:05:04 CET] Hi, how to know the version of a specific encoder within ffmpeg? (I want to check wich version of VP9 is used) [20:22:55 CET] hey anyone know if -af volume requires something compiled into ffmpeg ? i seem to be using the param, its not complaining about anything but .. the volume does not change [20:49:28 CET] I'm getting a new error trying to use the -filter_complex to concatenate movies into a single RTMP stream: vf/-af/-filter and -filter_complex cannot be used together for the same stream. [20:49:35 CET] I used -vf to resize the movies to 480p for the stream. [20:58:55 CET] Aaand? [21:20:18 CET] hi, what does "-b:v" means-do? where can i find the explanation of that on the help documents? [21:21:53 CET] autofsckk, it sets bitrate (-b) for video (:v) [21:22:24 CET] thanks Mavrik [21:24:09 CET] Mavrik -> do you know a good source of information on how to use maybe a script to put in all the things to be done to a conversion? im doing more filters and stuff to videos, but i find a little complicated to do it all in console, i read that it is possible to use an external file but cant find about it yet on the help how to do it [21:24:32 CET] Hrmf, no idea out of my head, perhaps try ffmpeg wiki? [21:27:00 CET] im amazed with ffmpeg, so powerful tool [22:04:20 CET] autofsckk: -filter_script script [22:07:47 CET] ok durandal_1707 thanks [22:29:21 CET] , video bit rate is what it means [22:50:04 CET] Hello [22:51:02 CET] Do you know how to allow external connection with ffserver ? [22:51:02 CET] I don't understand the doc [22:54:08 CET] Sorry lagg problem [23:01:16 CET] Phrk_: i think there is a configuration for access restriciton [23:01:21 CET] in the configuration file i mean [23:02:43 CET] what are you doing, what is happening and what are you expecting to happen instead? [23:08:32 CET] ffserver say : HTTP/1.1" 301 208 [23:08:37 CET] for my external ip [23:08:44 CET] but work for localhost [23:09:17 CET] the and the i don't understand how to allow everything [23:15:58 CET] * don't work [23:16:05 CET] klaxa [23:16:23 CET] can you pastebin your ffserver.conf? [23:17:50 CET] of course [23:18:42 CET] klaxa : http://hastebin.com/ifedimuroj.coffee [23:20:49 CET] Application error ? Sorry for that : http://pastebin.com/Gui33CmC [23:20:50 CET] uh... hmm... [23:21:06 CET] ah they fixed it [23:21:11 CET] no i'm just pondering what could be the issue [23:21:28 CET] how do you try to send the stream to the server? [23:21:49 CET] just ffmpeg -i somefile.mkv http://someserver/feed1.ffm ? [23:22:50 CET] 301 is redirection, maybe you are using the wrong url and it is trying to redirect you to ffmpeg.org? [23:23:35 CET] ah, are you trying to watch a stream or send a feed? [23:23:48 CET] Phrk_: ^ [23:24:47 CET] when i send to stream to localhost [23:24:53 CET] everything is working [23:25:12 CET] but when i try to read it with like vlc, mpv, or ffplay [23:25:22 CET] i can only read it from localhost, klaxa [23:25:41 CET] how do you try to read it? [23:26:01 CET] from lan : mpv http://127.0.0.1:8090/live.flv [23:26:05 CET] you should be able to watch http://someserver/live.flv [23:26:06 CET] yeah [23:26:07 CET] hmm... [23:26:22 CET] when i try from internet ffserver say something [23:26:24 CET] not sure how the ACL parts are parsed [23:26:32 CET] HTTP/1.1" 301 208 [23:27:00 CET] maybe they overwrite? (doubt it though) but maybe removing the superfluous ones (localhost and 192.168.178.25) could help? [23:27:17 CET] hmm.. but that really shouldn't be [23:28:00 CET] i'll try to use your config and see if i can reproduce [23:28:02 CET] ACL deny 1.0.0.0 1.255.255.255 would deny the whole of network 1 and allow everybody else. [23:28:12 CET] how i allow everybody [23:29:45 CET] hmm... [23:30:07 CET] address is not further specified, so i can't tell whether or not it accepts wildcards [23:30:23 CET] but ACL allow 0.0.0.0 255.255.255.255 should cover all addresses right? [23:33:25 CET] :( [GET] "/ HTTP/1.1" 301 208 [23:33:55 CET] :( [23:34:50 CET] what bind ? [23:34:58 CET] cause i have a lot of ethernet card [23:35:21 CET] if you get a connection it's not a networking error [23:35:28 CET] or at least it should not be :S [23:35:49 CET] ye of course [23:36:18 CET] Well time for bed, try this tommorow will be better [23:36:22 CET] gotta go, maybe i'll tinker with this later tonight [23:36:25 CET] Thanks [23:36:29 CET] kk [00:00:00 CET] --- Mon Dec 7 2015 From burek021 at gmail.com Tue Dec 8 02:05:02 2015 From: burek021 at gmail.com (burek) Date: Tue, 8 Dec 2015 02:05:02 +0100 (CET) Subject: [Ffmpeg-devel-irc] ffmpeg.log.20151207 Message-ID: <20151208010502.AA8B12AD6026@apolo.teamnet.rs> [00:34:23 CET] Hi relaxed, JEEB [00:34:30 CET] are you there? [01:03:20 CET] so what's a reasonable hls setting for low buffering? [01:03:33 CET] i've been doing -hls_time 3 -hls_list_size 10 [03:13:36 CET] Hey , relaxed? [03:14:06 CET] Relaxed? you have been relaxing so long time... :) [03:57:44 CET] ffmpeg won't authenticate to Flash Media Server. [04:00:51 CET] http://pastebin.com/vp54NZad [04:01:14 CET] hey guys [04:01:19 CET] ffmpeg question: I'm trying to publish to an FMS server. I get this output with wrong password. [04:01:39 CET] question... if i have an input that is not 16/9 and has black bars on each side.. is there some way for me to stretch it and remove the black bars of the video without transcoding it ? [04:01:58 CET] i don't want to cuase cpu power just to remove the black bars and strech the video the the absolute left and right [04:02:01 CET] any hint ? [04:02:18 CET] you can't unless you do it on the player [04:02:34 CET] I have it compiling in OK and running but I am getting an authentication [04:02:34 CET] error. [04:02:55 CET] I have also tried compiling with --enable -openssl. [04:03:09 CET] Platform Ubuntu 15.10 [04:03:42 CET] damn [04:31:39 CET] I compile ffmpeg and h264 separately with MSVC successfully but when I enable-lib246 --enable-gpl then error always occurs "ERROR: libx264 not found", but actually the x264 library has already been in PREFIX=/usr/ [04:31:54 CET] do you know why? [04:50:13 CET] why ffmpeg can't answer correctly this question? [05:29:07 CET] hey guys.. anyone nkow what -af volume=-10dB would not work ? [05:29:22 CET] for some reason its not working anymore.. i set volume like -whatever and still the volume is the same.. strange. [05:29:46 CET] prelude2004c: hallo [05:36:03 CET] nevermind, figured it out :) [05:39:24 CET] prelude2004c: das ist echt wundersch�n [05:50:51 CET] sorry i dont understand [06:24:31 CET] could ffmpeg make a "quadbox" from four separate input streams? as shown here: https://www.youtube.com/watch?v=xRpv3fO4MlY [06:29:10 CET] causa_sui: https://ffmpeg.org/ffmpeg-filters.html#tile [06:30:45 CET] anyone know how i can get vdpau going and decoding on my nvidia cards ? i don't use the UI as i plan on using this for ffmpeg decoding.. please help .. i been stuck for days [06:31:10 CET] actually nvm that's just for images [06:31:13 CET] causa_sui: https://trac.ffmpeg.org/wiki/Create%20a%20mosaic%20out%20of%20several%20input%20videos [06:32:08 CET] furq: thank you [07:05:54 CET] i see aac is finalized!! yes! [07:16:12 CET] I found a weird(?) bug in ffmpeg: if the input jpg images don't have the file extension, ffmpeg throws an error [07:16:46 CET] "Invalid data found when processing input" [07:40:51 CET] I thought aac was live already per twitter post? just tried -c:a aac and got -strict 2 error still [07:40:56 CET] boo [08:10:58 CET] hi [08:13:15 CET] why this gives me an error: ffmpeg -f dshow -i "2" -c:a libvo_aacenc -b:a 64k -ac 2 -y "\Users\user1\1.wav" [08:13:23 CET] Unknown input format: 'dshow' [08:19:13 CET] i figured [08:22:49 CET] droid909: hallo [08:46:20 CET] waressearcher2: hi [10:30:51 CET] droid909: wie geht's es dir ? [11:07:43 CET] waressearcher2: yup, thanks [12:07:27 CET] my kernel is 2.6.32. The lastest static ffmpeg cannot run [12:07:41 CET] FATAL: kernel too old [12:07:42 CET] Segmentation fault [12:08:12 CET] how to decide the sutable version for my centos6? [12:12:47 CET] build it yourself or upgrade your kernel [12:13:56 CET] JEEB: update kernel now [12:14:23 CET] Hi [12:15:36 CET] jimgray: me? I'm upgrading my kernel just fine :P [12:17:09 CET] JEEB: i said me [12:26:38 CET] Hi, I need to decode raw bitstream of 12bit monochrome video, where two pixels are encoded in three bytes. Pixel0[7:0] Pixel1[3:0] Pixel0[11:8] Pixel1[11:4]. Does current ffmpeg has support for such pixel format? [12:45:01 CET] Guest22966: hallo [12:45:07 CET] Phrk_: wie geth's ? [12:47:00 CET] someone know ffserver ? [12:47:11 CET] I don't understand how to accepte people from outside lan [12:47:20 CET] he say error http 301 [12:47:59 CET] klaxa tried to help me yersteday we don't find any solution [13:23:17 CET] Hello [13:23:53 CET] How to Convert any video in MP4 using ffmpeg. [13:24:14 CET] t4nk394: paste that last sentence without the final "." into google ... [13:24:51 CET] !g [13:25:03 CET] I have try with below command [13:25:21 CET] ffmpeg -i $input -s 320x240 -ar 44100 -r 12 $output [13:25:44 CET] but converted file now support in HTML5 video tag [13:25:59 CET] so - let's assume you want to understand what you are doing - what do you think does this command do? [13:26:43 CET] Any body have perfect solution [13:34:33 CET] t4nk394: ffmpeg -i 1.avi -c:v mpeg4 -vb 2000k -ab 320k -ar 44100 -ac 2 -c:a libmp3lame -y -f avi 2.avi [14:11:57 CET] Can the ffmpeg or ffplay commands be used to detect corruption in common file types (e.g. AVI, MOV, MPG, MKV, MTS)? [14:12:52 CET] I see that ffplay has an -err_detect option, but can ffplay just try to decode a file as fast as possible and not actually output video or audio? [14:13:09 CET] i think it show this in the log [14:16:14 CET] And I kinda need to use it in a script, just to check a large collection of videos after data recovery from a failed hard drive. [14:16:36 CET] ah sorry to much for me [14:16:44 CET] So it would be nice if ffmpeg can indicate corruption via exit code [14:16:54 CET] Did you search the doc ? [14:18:42 CET] k_sze: you can try with ffmpeg -f null but i have broken files which that doesn't flag up [14:29:39 CET] I use ffprobe -show_frames to decode the whole video. then watch for errors [14:30:12 CET] in terms of automating the process, that's up to you. you might also consider -print_format to select how the regular output is formatted [14:44:52 CET] ah, ffprobe also has the -show_error option. [14:46:50 CET] err, ffprobe -show_frames is so verbose. [14:47:17 CET] I don't need that kind of verbosity in a script. [14:47:32 CET] I guess I'll stick to ffmpeg [14:49:53 CET] `ffmpeg -v error -xerror -i foo.mkv -f null -` should do, I think. [14:50:12 CET] or even without the -xerror [14:51:27 CET] true, but it does force ffprobe to decode all frames. [14:51:39 CET] while not depending on any hardware and running as fast as the disk/CPU allows [14:56:53 CET] woohoo, I'm burning all 8 cores of my CPU. [14:57:26 CET] if only I know how to compile ffmpeg with nvenc on OS X. [14:57:49 CET] k_sze it's not that hard [14:58:03 CET] just use the basic compile way and add the good option [14:59:51 CET] --enable-nonfree \ [14:59:51 CET] --enable-nvenc \ [15:01:15 CET] Someone is good with ffserver here ? [15:06:05 CET] There is no nvenc on OSX. [15:07:23 CET] Not really convinced that it takes 4 cores to burn through 12 MB/s [15:11:39 CET] H264 is actually fairly hard on the CPU to decode. compared to, say, mpeg2 or xvid [15:11:53 CET] you can disable CABAC if you really want to [15:12:05 CET] if you are encoding the content that is [15:12:12 CET] but yeah, CABAC + a lot of bit rate can be funky [15:12:34 CET] would that risk any decoding errors? [15:12:54 CET] you can't skip it if the content is coded with CABAC :P [15:12:55 CET] oh wait... there really is no NVENC sdk for OS X. [15:12:57 CET] https://developer.nvidia.com/nvidia-video-codec-sdk#gpulist [15:12:58 CET] :( [15:13:17 CET] which is why I noted that it's the one alternative to have your content (that you encode) playable with a first-gen xbox [15:13:28 CET] ("if you are encoding the content that is") [15:17:04 CET] k_sze, on OSX you have to use their videotoolbox [15:17:56 CET] And it'll pass it on to the HW encoder [15:21:10 CET] Mavrik: videotoolbox?? [15:21:42 CET] Yes, the OS X / iOS video processing API. [15:22:01 CET] Successor to VDA [15:22:49 CET] VDA works as well (IIRC ffmpeg supports it) but it's rather buggy and decode only. [15:24:05 CET] And I suppose ffmpeg isn't taking advantage of that? [15:24:28 CET] I don't think the patch was merged. [15:24:28 CET] I mean, at least currently no build option would enable ffmpeg to take advantage of that. [15:48:25 CET] Any help with ffserver ? :( [16:19:27 CET] ask on ml? [17:29:40 CET] I'm trying to convert ismv (microsoft smoothstream, uses the 'mov,mp4,m4a,3gp,3g2,mj2' demuxer) to mpegts with transcoding. While it works, the output seems has bad PCR and only 1 PCR value comes in every 2 seconds. (The smoothstream chunks the video in 2s increments). Anyone know how to forcibly re-clock the video? [18:48:00 CET] so I encoded two files at 4000kbps. One in libx265 the other in libx264. x264 was 312MB, x265 was 303MB. Are the compressions they talk about x265 working? [18:48:27 CET] the 20%-50%? [18:50:19 CET] That's just the rate control being different [18:51:20 CET] i see [18:51:34 CET] this should force it to stick though yes? ffmpeg -i bbb_sunflower_2160p_30fps_normal.mp4 -c:v libx265 -preset veryslow -b:v 7501k -minrate 7501k -maxrate 7501k -an BBB_H265_VERY_SLOW_NOAUDIO_2160P_7501k.mkv [18:52:48 CET] you set a bitrate. both codecs will try to target that number, but without 2-pass mode there's going to be some error [18:53:14 CET] minrate and maxrate don't make a true cbr file, unfortunately. it just brings it a lot closer [18:53:32 CET] and the two pass method? [18:54:05 CET] 2-pass lets the codec see the whole video first, then it gets another chance to process it with information saved the first time around. bitrate tends to be much closer to the target [18:54:26 CET] not perfectly, but the variance might be closer to 2 megabytes [18:54:28 CET] i see [18:55:15 CET] so if the separation was 9MB. x264 v x265. If I do a 2-pass on x265, the file might get even smaller? [18:56:23 CET] not smaller. you set a target bitrate of 7501 kbit. which means you should be able to calculate the exact file size (muxing notwithstanding) [18:56:35 CET] afk [18:56:36 CET] on the documentation they used: /dev/null && \ in windows command... that is going to be different [19:03:08 CET] regarding my PCR issue, I think the issue is the PCR interval is initially calculated for the source material's timebase, but then timestamps start coming in with mpegts timebase so the PCR is way off kilter [19:17:36 CET] DHE: so 2pass will never get smaller files vs 1 pass? [19:18:12 CET] pinPoint: within the expected variance, setting the -b:v parameter will dictate the target file size [19:18:29 CET] 1pass or 2pass doesn't matter, the target size is exactly that [19:19:16 CET] ok [19:31:29 CET] Hey all, I am struggling getting ffmpeg working with NVENC support on ubunut. I've installed the cuda libraries from the nvidia sdk (they show up in ldconfig -p), compiled ffmpeg with --enable-nvenc, yet ffmpeg still keeps telling me "Failed loading CUDA library" when I pass in -vcodec nvenc. Any ideas? [19:32:48 CET] Looking at the ffmpeg source it appears that message occurs when trying to load the libcuda.so, but since ldconfig -p shows that library I'm unsure of what the actual issue is [20:23:51 CET] is YUVY just YUV 422? [20:40:43 CET] another way of holding it, yes [20:41:02 CET] the data is the same after you parse it, but it's packed differently [20:41:27 CET] Well turns out I have YUVY/I420 either or [20:41:40 CET] and I need it in a opencv Mat class :) [20:42:09 CET] lol [20:46:36 CET] Mat imgMat = Mat(height * 1.5f, width, CV_8UC1, data); i hope that works [20:46:54 CET] no that won't [20:46:58 CET] it's just one layer [20:47:03 CET] CV_8UC1 [20:47:11 CET] 0..255 byte image [20:47:21 CET] ah now I got it [20:47:30 CET] yeah, you might get that stored in there [20:47:40 CET] but opencv cannot really know what to do with it [20:55:51 CET] what did you say [20:55:56 CET] before i lost the internet [20:56:27 CET] will i need to call cvtColor(myuv [21:05:32 CET] come back to me! [22:25:13 CET] hey , anyone see something wrong with this : http://pastebin.com/kzA96Znt .. i am seeing an error : Enter command: |all