[Ffmpeg-devel-irc] ffmpeg-devel.log.20111228

[burek]

[02:21] <CIA-49> ffmpeg: 03Michael Niedermayer 07release/0.6 * re1a2bcbec8 10ffmpeg/libavcodec/h264.h: 
[02:21] <CIA-49> ffmpeg: h264: fix init of topleft ref/mv.
[02:21] <CIA-49> ffmpeg: Fixes Ticket778
[02:21] <CIA-49> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[02:21] <CIA-49> ffmpeg: (cherry picked from commit 680880c98db2817437e19c3fc7f6349261bbbbb0)
[02:21] <CIA-49> ffmpeg: 03Michael Niedermayer 07release/0.7 * r575cbbffaa 10ffmpeg/libavcodec/h264.h: 
[02:21] <CIA-49> ffmpeg: h264: fix init of topleft ref/mv.
[02:21] <CIA-49> ffmpeg: Fixes Ticket778
[02:21] <CIA-49> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[02:21] <CIA-49> ffmpeg: (cherry picked from commit 680880c98db2817437e19c3fc7f6349261bbbbb0)
[02:21] <CIA-49> ffmpeg: 03Tom Evans 07release/0.9 * rf7c196a1f9 10ffmpeg/ (configure libavdevice/v4l2.c): 
[02:21] <CIA-49> ffmpeg: libavdevice: Fix v4l2 compilation under FreeBSD
[02:21] <CIA-49> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[02:21] <CIA-49> ffmpeg: (cherry picked from commit ffa1f541b350946591ea80e134917b28a860fcad)
[02:51] <CIA-49> ffmpeg: 03Martin Storsjö 07master * r676a9ee1d2 10ffmpeg/libavcodec/ (h264_cabac.c x86/h264_i386.h): (log message trimmed)
[02:51] <CIA-49> ffmpeg: x86: Fix constraints for decode_significance*_x86
[02:51] <CIA-49> ffmpeg: Originally, prior to 8742a4ff8, the caller code was compiled
[02:51] <CIA-49> ffmpeg: within this condition:
[02:51] <CIA-49> ffmpeg: ARCH_X86 && HAVE_7REGS && HAVE_EBX_AVAILABLE && !defined(BROKEN_RELOCATIONS)
[02:51] <CIA-49> ffmpeg: Since HAVE_7REGS is defined as
[02:51] <CIA-49> ffmpeg: (ARCH_X86_64 || (HAVE_EBX_AVAILABLE && HAVE_EBP_AVAILABLE))
[02:51] <CIA-49> ffmpeg: 03Janne Grunau 07master * r1adba23cec 10ffmpeg/libavdevice/v4l2.c: v4l2: remove unneded linux specific asm/types.h include
[02:51] <CIA-49> ffmpeg: 03Janne Grunau 07master * r2702ec2ef4 10ffmpeg/doc/doxy/ (footer.html header.html): doxygen: remove unreplaced variables from custom header and footer
[02:51] <CIA-49> ffmpeg: 03Aneesh Dogra 07master * rbd96be6e27 10ffmpeg/libavcodec/mpegvideo_enc.c: 
[02:51] <CIA-49> ffmpeg: mpegvideo_enc: K&R cosmetics
[02:51] <CIA-49> ffmpeg: Signed-off-by: Ronald S. Bultje <rsbultje at gmail.com>
[02:51] <CIA-49> ffmpeg: 03Michael Niedermayer 07master * r0e5fbbd776 10ffmpeg/: (log message trimmed)
[02:51] <CIA-49> ffmpeg: Merge remote-tracking branch 'qatar/master'
[02:52] <CIA-49> ffmpeg:  mpegvideo_enc: K&R cosmetics
[02:52] <CIA-49> ffmpeg:  doxygen: remove unreplaced variables from custom header and footer
[02:52] <CIA-49> ffmpeg:  threads: test for sys/param.h and include it for sysctl on OpenBSD
[02:52] <CIA-49> ffmpeg:  v4l2: remove unneded linux specific asm/types.h include
[02:52] <CIA-49> ffmpeg: 03Janne Grunau 07master * r06bb590558 10ffmpeg/ (configure libavcodec/pthread.c): threads: test for sys/param.h and include it for sysctl on OpenBSD
[06:01] <CIA-49> ffmpeg: 03Carl Eugen Hoyos 07master * r190a0998c3 10ffmpeg/libavcodec/pnm.c: 
[06:01] <CIA-49> ffmpeg: Fix a crash when reading gray pam files.
[06:01] <CIA-49> ffmpeg: Fixes ticket #837.
[06:01] <CIA-49> ffmpeg: 03Carl Eugen Hoyos 07master * rc1bc1967f3 10ffmpeg/libavcodec/ (pamenc.c pnm.c): 
[06:01] <CIA-49> ffmpeg: Fix a typo that broke writing (and reading) PAM files.
[06:01] <CIA-49> ffmpeg: Reported and reviewed by Derek Buitenhuis.
[06:02] <Daemon404> finding that bug was just a fluke 
[06:02] <Daemon404> photoshop and gimp cant even open true pam files anyway
[06:02] <Daemon404> lol
[08:31] <GorDonFreeMan>     return (t1.tv_sec - t2.tv_sec) + (t1.tv_nsec - t2.tv_nsec) * (double_t)SEC_IN_NSEC;
[08:31] <GorDonFreeMan> is this the fastest way i can do it?
[08:32] <GorDonFreeMan> /* static inline double_t calcdiff_double(struct timespec t1, struct timespec t2) */
[13:23] <ubitux> hi saste, "long" time no see :)
[13:27] <saste> ubitux: yep, i get even lazier during festivities
[17:15] <ubitux> wtf are the negative timecode in the mov?
[17:16] <av500> watch it backwards....
[17:16] <ubitux> huh?
[17:45] <ubitux> av500: what did you mean? was it a joke? :p
[17:45] <gnafu> ubitux: Coming from him, I'm going to say, "Yes."
[17:45] <gnafu> ;D
[17:46] <ubitux> :(
[17:46] Action: gnafu hugs ubitux.
[17:46] Action: ubitux feels much better now
[17:46] <ubitux> this is pretty weird
[17:46] <ubitux> this is what the specs say: http://pastie.org/3084906
[17:46] <ubitux> "A 1-bit value indicating the times sign. If bit is set to 1, the timecode record value is negative."
[17:47] <av500> timecode?
[17:47] <nevcairiel> the spec is annoying, it always says what something is, not why that is =p
[17:47] <av500> most specs are like that
[17:48] <kierank> weird, negative timecode
[17:48] <ubitux> kierank: yes, i don't understand what it means&
[17:48] <kierank> negative timecode i guess
[17:48] <kierank> =p
[17:48] <ubitux> ah ok, thanks
[17:48] <ubitux> ;)
[17:48] <kierank> feel free to implement timecode in x264
[17:48] <kierank> with all the weird pulldown combinations
[17:48] <kierank> =p
[17:48] <av500> timecode as in "not timestamp"?
[17:48] Action: michaelni wants (5h 53min)*sqrt(-1)
[17:49] <ubitux> complex time
[17:49] <kierank> timecode in reference frame x
[17:51] <ubitux> timecode ’ frame, and backward frame if timecode<0 ?
[17:51] <ubitux> :/
[17:52] <ubitux> anyway, the timecode seems to be stored as a basic frame number in the sample i have
[17:52] <ubitux> and doesn't seem to match the actual layout described
[17:53] <ubitux> maybe the file isn't a mov but a mp4, and the specs differ in that matter
[17:54] <ubitux> < kierank> with all the weird pulldown combinations // this reminds me ffmepg is lacking a "pullup" filter
[17:54] <kierank> vlc has one =p
[17:55] <ubitux> yes, and mplayer has at least 3
[17:55] <ubitux> seems there are 4
[17:55] <ubitux> detc, ivtc, pullup and filmdint
[18:01] <relaxed> ffmpeg has three of the above via the mp filter
[18:01] <relaxed> unless I'm missing something
[18:01] <ubitux> do they actually work?
[18:01] <ubitux> i think there is an issue with one of them
[18:02] <relaxed> I haven't tested them but they're listed under mp in the docs.
[18:02] <ubitux> https://ffmpeg.org/trac/ffmpeg/ticket/681
[19:21] <CIA-49> ffmpeg: 03Michael Niedermayer 07master * r11ca3416f9 10ffmpeg/libavcodec/v410dec.c: 
[19:21] <CIA-49> ffmpeg: v410dec: Check for sufficient input data.
[19:21] <CIA-49> ffmpeg: Fixes crash
[19:21] <CIA-49> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[19:22] <CIA-49> ffmpeg: 03Michael Niedermayer 07master * r8e09482e4d 10ffmpeg/libavcodec/vcr1.c: 
[19:22] <CIA-49> ffmpeg: vcr1dec: Check that there is sufficient input data.
[19:22] <CIA-49> ffmpeg: Fixes crash.
[19:22] <CIA-49> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[20:40] <CIA-49> ffmpeg: 03Michael Niedermayer 07master * r4b35ee0b7c 10ffmpeg/libavcodec/ivi_common.c: 
[20:40] <CIA-49> ffmpeg: indeo5: fix null pointer crash with ref_mb
[20:40] <CIA-49> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[20:40] <CIA-49> ffmpeg: 03Michael Niedermayer 07master * rd46bc4133c 10ffmpeg/libavcodec/indeo5.c: 
[20:40] <CIA-49> ffmpeg: indeo5: Fix crash due to partially initialized gop vars.
[20:40] <CIA-49> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[21:59] <Daemon404> man trac is slow
[21:59] <cbsrobot> so right
[22:00] <gnafu> Yet so wrong.
[22:00] <Daemon404> if it's anything like aegisub's trac
[22:00] <Daemon404> it's running like 10000 SQL queries to bring up one bug page
[22:26] <CIA-49> ffmpeg: 03Michael Niedermayer 07master * r92e2b59dec 10ffmpeg/libavcodec/ivi_common.c: 
[22:26] <CIA-49> ffmpeg: indeo5: fix division by 0 in ff_ivi_init_tiles()
[22:26] <CIA-49> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[22:26] <CIA-49> ffmpeg: 03Michael Niedermayer 07master * ra1ca45317d 10ffmpeg/ (3 files in 3 dirs): 
[22:26] <CIA-49> ffmpeg: cljrenc: fix scaling
[22:26] <CIA-49> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[22:26] <CIA-49> ffmpeg: 03Michael Niedermayer 07master * r7e5cbb3c2d 10ffmpeg/ (4 files in 4 dirs): 
[22:26] <CIA-49> ffmpeg: cljrdec: improve scaling somewhat
[22:26] <CIA-49> ffmpeg: Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[23:10] <ffmpeguser> hi all, is the code that changes the aspect ratio broken? i tried with a git build that is only about 2 days old and no matter what parameters i use the aspect ratio of the video stream that i want to change in the output is always the same
[23:14] <funman> ffmpeguser: i think you are on devel channel, not user
[23:14] <ffmpeguser> i know, and only the developers can answers the question
[23:15] <ffmpeguser> answer*
[23:15] <Daemon404> please ask in #ffmpeg
[23:15] <Daemon404> it sounds liek a usage problem for sure
[23:18] <cbsrobot> ffmpeguser: I gave you the answer on #ffmpeg already
[23:18] <ffmpeguser> ?
[23:18] <ffmpeguser> the last thing you said 'well..'
[23:19] <ffmpeguser> said was*
[23:20] <cbsrobot> [22:24] ffmpeguser: hi all, is it possible to remux a video stream with -vcodec copy, changing only the aspect ratio of it?
[23:20] <cbsrobot> [22:33] cbsrobot: ffmpeguser: Idont think so - but I'm not sure - try it out
[23:21] <ffmpeguser> but i want to ask one more thing besides that, "[alac @ 003EA2A0] Unsupported channel count: 6" is support for that being planned?
[23:21] <ffmpeguser> we tried everything cbsrobot, the aspect ratio in the output file was always the same
[23:22] <cbsrobot> as I said: with -vcodec copy the -vf has no effect
[23:22] <Daemon404> ffmpeguser, what input/output
[23:22] <Daemon404> does it happen to be .mov?
[23:23] <ffmpeguser> but i tried without -vcodec also and the results were always the same
[23:23] <Daemon404> and are you trying to set it soft
[23:23] <ffmpeguser> .flv
[23:23] <Daemon404> and teh output?
[23:23] <ffmpeguser> i'll copy-paste
[23:26] <ffmpeguser>     Stream #0:0(eng): Video: h264 (Main) (avc1 / 0x31637661), yuv420p, 720x390, 1564 kb/s, 25 fps, 25 tbr, 2500 tbn, 5k tbc
[23:27] <ffmpeguser> ffmpeg.exe -i "file.flv" -aspect 10:10 test.flv
[23:28] <ffmpeguser> the process starts and i stop it after a few seconds with Ctrl+C and i check the file with ffplay
[23:29] <ffmpeguser> Stream mapping:   Stream #0:0 -> #0:0 (h264 -> flv)   Stream #0:1 -> #0:1 (aac -> libmp3lame)
[23:30] <ffmpeguser> -aspect, -setdar, -setsar we tried all three one by one, the result is always the same
[23:30] <cbsrobot> Not sure flv even supports aspect ratio
[23:31] <ffmpeguser> i'll try avi
[23:34] <ffmpeguser> it did change it
[23:34] <ffmpeguser> mkv too
[23:35] <ffmpeguser> then the question is this, will it be possible in the future to change it with with -vcodec copy set?
[23:35] <ffmpeguser> -with
[23:39] <ffmpeguser> are those 2 things being worked on?
[23:42] <Daemon404> um
[23:42] <Daemon404> i dont think flv even has an aspect ratio tag.
[23:42] <Daemon404> so it's not possible to set it.
[23:42] <Daemon404> you'd have to hard-resize
[23:42] <Daemon404> (or if it does, nothing supports it)
[23:42] <Daemon404> it's a container limitation i believe.
[23:44] <ffmpeguser> i used flv, when i tested all that but there is no problem if mkv for example, it's a good container format
[23:44] <ffmpeguser> mkv is used*
[23:45] <ffmpeguser> hard-resize?
[23:45] <Daemon404> as in you have to actually resize the image
[23:46] <Daemon404> to a new shape
[23:46] <Daemon404> what youre doing is trying to set teh container flag
[23:46] <ffmpeguser> hm
[23:52] <Shimmy> Anyone please help me: https://ffmpeg.org/trac/ffmpeg/ticket/747
[23:52] <Shimmy> Pleeeeeeeeeease?
[23:56] <ffmpeguser> i'll checking to learn if i can change that with mkvtoolnix
[23:56] <ffmpeguser> i'm*
[00:00] --- Thu Dec 29 2011