[Ffmpeg-devel-irc] ffmpeg-devel.log.20140911

[burek]

[01:05] <Timothy_Gu> See if they respond to my last question: https://lists.libav.org/pipermail/libav-devel/2014-September/063150.html
[01:24] <Timothy_Gu> kierank: the link on http://obe.tv/ to your Opus blog post is dead
[01:25] <Timothy_Gu> kierank: also the second "read more" on http://obe.tv/products
[01:33] <cone-182> ffmpeg.git 03Diego Biurrun 07master:775a0b04f0cf: doc: Fix syntax and logical errors in avconv stream combination example
[01:33] <cone-182> ffmpeg.git 03Michael Niedermayer 07master:5e3da256184e: Merge commit '775a0b04f0cf8102fe322b2ee03fe1a0633dea04'
[01:40] <wm4> why is this linked on ffmpeg.org http://www.mplayerhq.hu/DOCS/tech/patches.txt
[01:41] <wm4> that's negatively helpful
[01:41] <wm4> starting with the fact that it assumes svn
[01:41] <J_Darnley> because the docs are old
[01:42] <wm4> anyway, this guy has a point: https://ffmpeg.org/pipermail/libav-user/2014-September/007400.html
[02:13] <Timothy_Gu> wm4: link to the MPlayer doc?
[02:13] <wm4> Timothy_Gu: see above, if you mean where it's linked: http://ffmpeg.org/contact.html#MailingLists
[02:14] <Timothy_Gu> wm4: that's what i meant
[02:14] <Timothy_Gu> interresting
[05:12] <cone-1> ffmpeg.git 03Michael Niedermayer 07master:16de4d927e8b: ffmpeg: Do av_buffersink_set_frame_size() when reconfiguring the filtergraph not just when changing audio resample parameters
[11:43] <rcombs> wait a second, why am I allowed to close or assign other people's tickets
[11:47] <ubitux> because you opened it maybe
[11:48] <rcombs> I get the same options on all tickets, though
[11:48] <ubitux> are you going to use that power for evil?
[11:48] <rcombs> I'd expect not
[11:49] <rcombs> "oh no, the evil rcombs is pillaging the town and marking tickets as solved!"
[11:49] <rcombs> "and assigning ASS-related bugs to wm4!"
[11:49] <ubitux> ahah
[11:49] <ubitux> LGTM
[11:58] <ubitux> we should add harfbuzz support to drawtext :-°
[11:59] <rcombs> mpv actually implements its OSD using libass
[12:08] <rcombs> ^ does that happen often?
[12:22] <ubitux> rcombs: there is a word blacklist somewhere
[12:23] <ubitux> it used to happen often
[12:24] <ubitux> eeeh no escaping possible with file with concat demuxer :(
[14:13] <cone-520> ffmpeg.git 03Michael Niedermayer 07master:ea7ebadeb78c: avformat/rmdec: very basic MLTI support
[15:25] <cone-520> ffmpeg.git 03Michael Niedermayer 07master:5303a644f588: avformat/img2: Make AVOptions available to img* demuxers defined in other files
[15:25] <cone-520> ffmpeg.git 03Michael Niedermayer 07master:646d663f7137: avformat/img2_brender_pix: Add AVClass and image2 AVOptions
[15:25] <cone-520> ffmpeg.git 03Michael Niedermayer 07master:8b4714614f8b: avformat/img2_alias & brender_pix: fix function names
[16:54] <cone-520> ffmpeg.git 03Michael Niedermayer 07master:de4a4ca1a08d: avformat/img2_alias_pix: Add AVClass and image2 AVOptions
[16:54] <cone-520> ffmpeg.git 03Michael Niedermayer 07master:a6153bde3dee: avcodec/pngdec: print the actual number of bytes left in the error case
[16:54] <cone-520> ffmpeg.git 03Michael Niedermayer 07master:4a203b773702: avcodec/png_parser: Return the buffered data at the end
[17:03] <ubitux> > Is it possible for me to use part of this patch for the FFmpeg website?
[17:03] <ubitux> *sigh* - once more this feels like FFmpeg takes, takes, takes, and gives
[17:03] <ubitux> nothing in return (let's not even delve into the whole negativity thing).
[17:03] <ubitux> lol
[17:06] <rcombs> wat
[17:07] <ubitux> rcombs: yeah well they are upset because we steal everything from libav
[17:08] <rcombs> :|
[17:09] <ubitux> and here one guy was reviewing stuff from libav, and asked for the permission to pick a diff (which happens to be unlicensed)
[17:09] <ubitux> but oh well :)
[17:26] <ubitux> Timothy__Gu: stop stealing from libav!!
[17:30] <Timothy__Gu> ubitux: I can only sigh
[17:53] <J_Darnley> Woo!  A complex visualization rendering at more than realtime!
[17:53] <ubitux> ?
[17:55] <J_Darnley> The vis filter I've been working on.
[17:55] <ubitux> ok :)
[17:56] <ubitux> can you spoil a bit?
[17:58] <J_Darnley> Spoil a bit?
[17:58] <ubitux> tell us more
[17:58] <Daemon404> printf("3+2i");
[17:58] <Daemon404> compelx visualization!
[17:58] <J_Darnley> :)
[17:59] <J_Darnley> It is the port of Winamp's/Nullsoft's Advanced Visualization Studio I've been working on for a couple of months now.
[18:01] <Daemon404> ah... what was their stuff called... it was somethign silly
[18:01] <Daemon404> like MILK
[18:01] <kepstin-laptop> wow, been a while since i've used that :)
[18:01] <kepstin-laptop> no, MILK was the successor to AVS
[18:01] <J_Darnley> There was also Milkdrop
[18:01] <Daemon404> ah ok.
[18:01] <Daemon404> its been a while
[18:01] <Daemon404> ;)
[18:01] <kepstin-laptop> J_Darnley: any chance you hung out in #avschat back in the day?
[18:01] <J_Darnley> That's all on the GPU I think.
[18:01] <J_Darnley> No
[18:01] <kepstin-laptop> ah, pity.
[18:01] <ubitux> or projectM maybe
[18:04] <J_Darnley> I've been using Lua to replace the original expression library
[18:04] <J_Darnley> and now I've stopped compiling the same expression over and over, and moved to LuaJIT
[18:05] <J_Darnley> It now renders one preset at 640x480 at 60Hz at 39 fps.
[18:06] <J_Darnley> Uh that should be 640x480 at 30Hz
[18:08] <J_Darnley> I should put a video up somewhere.
[18:37] <cone-520> ffmpeg.git 03Rong Yan 07master:8e8db16a2cf5: avutil/ppc/float_dsp_altivec: ppc: Fix bug in ff_vector_fmul_add_altivec() affecting little endian environments on POWER8
[18:50] <cone-520> ffmpeg.git 03Jörg Krause 07master:453944cc77b9: configure: fix check_cmd in check_host_cpp
[20:03] <llogan> rcombs: you don't have to worry about closing those spam tickets. one of us (michael, carl, alexander, or me) will eventually banish it
[20:04] <llogan> unfortunatly they seem to appear when I'm afk the majority of the time
[20:04] <rcombs> llogan: yeah, figured; saw it and went "ooh, something to kill"
[20:04] <llogan> it is satisfying to nuke it
[20:06] <llogan> now if i weren't so bad at regex...
[20:06] <rcombs> http://regexcrossword.com <-- play this for a while and you won't be
[20:10] Action: llogan button mashes
[20:12] <ubitux> why do i have the feeling that someone is probably doing exactly the same thing at the same moment when i'm writing some code for ffmpeg?
[20:13] <ubitux> and like, i absolutely need to hurry up before he posts that patch(set)
[20:18] <llogan> ubitux: http://www.youtube.com/watch?v=7YvAYIJSSZY
[20:21] <ubitux> :(
[20:46] <llogan> how was VDD?
[20:46] <llogan> oh, wait wrong year
[20:46] <llogan> duh, 20-21 sept
[20:47] <llogan> i have a problem of not thinking before typing
[20:48] <wm4> libav 11 still not released?
[20:48] <wm4> it's always the same shit...
[20:50] <nevcairiel> told you so
[21:13] <llogan> damn it. i didn't see that the attachment was 14 MB.
[21:15] <llogan> i wish mailman would bark about file size instead of unsubscribed when both are true.
[21:19] <ubitux> ass expert (wm4, rcombs?); any opinion on the ass_set_font_scale() patch on libavfilter/vf_subtitles.c?
[21:19] <wm4> ubitux: the first patch looks very wrong
[21:19] <ubitux> (https://ffmpeg.org/pipermail/ffmpeg-devel/2014-September/162755.html)
[21:19] <wm4> I'll reply
[21:19] <ubitux> thank you
[21:21] <ubitux> kind related, i'm probably going to add the libass hinting option btw, just like shaper
[21:27] <ubitux> mmh we don't set the pixel_aspect
[22:33] <J_Darnley> llogan: I wonder whether you should announce that we are dropping those old patch guidelines.
[23:01] <llogan> J_Darnley: i just thought it was confusing to users and nobody reads it anyway
[23:02] <J_Darnley> That's certainly true
[23:02] <J_Darnley> I'm just reading the git howto for the first time.
[23:03] <J_Darnley> (If it wasn't clear I was mostly joking about making an announcement)
[23:03] <wm4> another suggestion: make it clear on the website that ffmpeg-devel is for patches
[23:03] <wm4> a user was confused where he should send his patch to (he ended up on libav-user)
[23:04] <J_Darnley> I would phrase that as "ffmpeg-devel is the place to send patches"
[23:04] <J_Darnley> Otherwise it sounds like it should be nothing but patches.
[23:05] <llogan> wm4: yeah, that's what I was going to do after dropping this MPlayer patch (if I can get support to do so).
[23:05] <iive> patches and development of ffmpeg itself.
[23:05] <llogan> s/patch/link
[23:05] <wm4> nice
[23:06] <llogan> in general I want to reduce verbosity and try to make it more clear to users
[23:07] <J_Darnley> I support your change and I'll send an email shortly
[23:07] <llogan> thanks!
[23:08] <J_Darnley> Oh I see Carl has sent a reply
[23:48] <J_Darnley> We need to stop what ever transforms our docs into html turning two hyphens into an em dash
[23:52] <cone-520> ffmpeg.git 03Michael Niedermayer 07release/2.1:6f6b1d0ddb26: avformat/oggparseopus: factor opus_duration() out
[23:53] <cone-520> ffmpeg.git 03Michael Niedermayer 07release/2.1:64908f70e436: avformat/oggparseopus: calculate pts/dts for initial packets after seeking
[23:57] <kepstin-laptop> re closed ticket - due to the change in beheviour with the ffmpeg tool exit code in "errors", 2.2 isn't actually a compatible series with 2.1 :/
[00:00] --- Fri Sep 12 2014