December 2005 - ffmpeg-devel - lists.ffmpeg.org

what's the audio stream format decode by avcodec_decode_audio?
by sigsegv11＠gmail.com 09 Jan '06

09 Jan '06

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Is it depend on the source movie? If no, what's the audio stream format? thanks! - -- - - sigsegv -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) iD8DBQFDl8mbzm46SfqoYmQRAqxjAJ902jaEODrEogj59ygyVINcDgqL9wCfWe8p jd8p39sPzMD0LRhPQhwbOTk= =yg/6 -----END PGP SIGNATURE-----

3 2

[PATCH] from DivX, Part 1: cosmectic changes
by slhomme＠divxcorp.com 08 Jan '06

08 Jan '06

Hi everyone, This is the first patch from DivX in order to make our code as similar as the official FFMPEG as possible. This patch includes only the cosmetic changes so that after that we can proceed with the real juice (otherwise the noise from these changes is too big to find the actual fixes). It has : - tab removal (yes the official code has tabs) - use av_log instead of fprintf - missing includes (help MSVC at least and won't hurt anyone) - a few #ifdef to hide code when it's not enabled in the main config (to avoid having to fix MSVC portability issues) - more PRId64, PRIx64, etc Have fun.

10 34

ffmpeg doesn't compile on Mac OS X anymore...
by patrice.bensoussan＠gmail.com 05 Jan '06

05 Jan '06

Hello, It looks like there are two issues when trying to compile on Mac OS X Tiger : 1/ When using gcc 3.3, I reach the linking stage but get : /Users/patriceb/tmp/WMA/cvs-HEAD/ffmpeg/libavutil/bswap.h:93: warning: `bswap_16' defined but not used cc -Wl,-dynamic,-search_paths_first -dynamiclib -Wl,-single_module,- undefined,dynamic_lookup -flat_namespace -Wl,-install_name,/usr/local/ lib/vhook/null.dylib -g -o null.dylib null.o ld: flag: -undefined dynamic_lookup can't be used with MACOSX_DEPLOYMENT_TARGET environment variable set to: 10.1 /usr/bin/libtool: internal link edit command failed make[1]: *** [null.dylib] Error 1 2/ When using gcc 4.0, compilation aborts with the error (this is only a warning with gcc 3) make -C libavutil all cc -g -Wall -Wno-switch -no-cpp-precomp -pipe -fomit-frame-pointer - force_cpusubtype_ALL -Wno-sign-compare -faltivec -Os -mdynamic-no-pic -DHAVE_AV_CONFIG_H -I.. -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE - D_GNU_SOURCE -c -o mathematics.o mathematics.c In file included from mathematics.c:24: common.h:551: error: static declaration of 'lrintf' follows non- static declaration make[1]: *** [mathematics.o] Error 1 make: *** [lib] Error 2 Are these known issue? Thanks. Patrice

4 6

[PATCH] H.263: More descriptive error message
by tom-lists＠jambon.ll.cx 04 Jan '06

04 Jan '06

I think it would be nice if ffmpeg had a more friendly and less confusing error message when h263 is told to deal with an invalid picture size. The patch below is a one-liner. Thanks, --Tom Joseph --- mpegvideo.c.old 2005-12-13 22:56:55.000000000 -0500 +++ mpegvideo.c 2005-12-13 23:00:02.000000000 -0500 @@ -1138,7 +1138,7 @@ break; case CODEC_ID_H263: if (h263_get_picture_format(s->width, s->height) == 7) { - av_log(avctx, AV_LOG_INFO, "Input picture size isn't suitable for h263 codec! try h263+\n"); + av_log(avctx, AV_LOG_INFO, "The specified picture size of %dx%d is not valid for the H.263 codec. Valid sizes are 128x96, 176x144, 352x288, 704x576, and 1408x1152.\n", s->width, s->height); return -1; } s->out_format = FMT_H263;

2 3

bash substitutions
by jakemsr＠jakemsr.com 03 Jan '06

03 Jan '06

revision 1.217 of configure added some shell substitutions. these apparently only work with bash, which is _not_ the /bin/sh on all systems. specifically: echo "SPPMAJOR=${lavc_version/.*/}" >> config.mak echo "LAVCMAJOR=${lavc_version/.*/}" >> config.mak echo "LAVFMAJOR=${lavf_version/.*/}" >> config.mak echo "LAVUMAJOR=${lavu_version/.*/}" >> config.mak this not work with pdksh, which is the /bin/sh on OpenBSD. I doubt it works with ash, which is the /bin/sh on NetBSD. I suggest the following patch. -- <jakemsr at jakemsr.com> --- configure.orig Sat Dec 17 15:00:08 2005 +++ configure Sat Dec 17 15:26:44 2005 @@ -1457,13 +1457,13 @@ lavu_version=`grep '#define LIBAVUTIL_VE if test "$lshared" = "yes" ; then echo "BUILD_SHARED=yes" >> config.mak echo "PIC=-fPIC -DPIC" >> config.mak - echo "SPPMAJOR=${lavc_version/.*/}" >> config.mak + echo "SPPMAJOR=$(echo $lavc_version | sed 's/\..*//')" >> config.mak echo "SPPVERSION=$lavc_version" >> config.mak - echo "LAVCMAJOR=${lavc_version/.*/}" >> config.mak + echo "LAVCMAJOR=$(echo $lavc_version | sed 's/\..*//')" >> config.mak echo "LAVCVERSION=$lavc_version" >> config.mak - echo "LAVFMAJOR=${lavf_version/.*/}" >> config.mak + echo "LAVFMAJOR=$(echo $lavf_version | sed 's/\..*//')" >> config.mak echo "LAVFVERSION=$lavf_version" >> config.mak - echo "LAVUMAJOR=${lavu_version/.*/}" >> config.mak + echo "LAVUMAJOR=$(echo $lavu_version | sed s'/\..*//')" >> config.mak echo "LAVUVERSION=$lavu_version" >> config.mak fi echo "EXTRALIBS=$extralibs" >> config.mak

6 8

[PATCH] DV fix to enable mutliple run in the same process
by slhomme＠divxcorp.com 03 Jan '06

03 Jan '06

Everything is in the title.

2 1

[PATCH] from DivX, Part 12: DV aspect ratio
by slhomme＠divxcorp.com 03 Jan '06

03 Jan '06

This code moves the AR detection in the codec part so that it works when the DV stream is in AVI (or else) too. Plus the interlacing detection now works.

3 2

Re: TrueMotion decoder
by kostya.shishkov＠gmail.com 03 Jan '06

03 Jan '06

Now it has sane variable names, got rid of shorts and fixed typo spotted by Benjamin.

3 5

improving encoding (possibly big perceptual gains)
by dalias＠aerifal.cx 02 Jan '06

02 Jan '06

with all the recent discussions over improving lavc, i've got some ideas. mainly i'm worried that psnr is very misleading when trying to improve motion estimation decisions since often a particular motion vector will give better psnr but much worse actual quality due to loss of small detail and mud/blocking. because of this, i'd like to suggest that lavc be extended to measure the 'uqi' (universal [image] quality index) proposed by the paper at this site: http://www.cns.nyu.edu/~zwang/files/research/quality_index/demo_lena.html i think the images there speak for themselves regarding how stupid psnr is as a judge of quality. however, printing statistics isn't much use if you can't actually improve encoding. so the main idea is too add a *cmp function to measure the uqi and use this for motion estimation and optimal quantization and qprd, etc. maybe also the ratecontrol engine itself. this should do a much better job of telling when it's ok to use poor quantization, zero ac coeffs, etc. and when it will look like shit (like the last lena picture ;). the next more ambitious idea is for future codecs, perhaps snow: to use only the correlation part of uqi to evaluate motion vectors, then measure the difference in mean and variance and store those in the bitstream. this would allow complex changes such as fades and changes in lighting to be encoded almost entirely with motion vectors, whereas mpeg4 codecs usually use dct-coded residue for such scenes with horrible quality as the result unless qp is 2 or less. ideas, anyone? sorry i'm too lame to code on lavc.. actually i don't know the codebase and i'm busy working on another project right now. however the *cmp system seems extensible enough that adding a uqi method to it should be easy. the other things (quantization, qprd, rate control) might be more difficult. rich

4 5

[PATCH] second attempt to auto_guess streams in mpegts
by nsabbi＠tiscali.it 02 Jan '06

02 Jan '06

This patch permits to recognize some more mpegts samples: the ones that contain neither SDT nor PAT/PMT (although they are hardly mpegts). It simply tries to rescan the content with ts->auto_guess=1 if the first iteration fails. P.S. sometimes it's necessary to analyze a large amount of data to find something useful (e.g. not encrypted streams), even some MB. Is there a parameter to identify such "probing range"? Bye, Nico

3 2