[FFmpeg-trac] #1351(avcodec:reopened): mpeg2video lowres support
FFmpeg
trac at avcodec.org
Sat May 26 04:37:34 CEST 2012
#1351: mpeg2video lowres support
-------------------------------------+-------------------------------------
Reporter: jyavenard | Owner:
Type: defect | Status: reopened
Priority: important | Component: avcodec
Version: git-master | Resolution:
Keywords: mpeg2video | Blocked By:
regression lowres | Reproduced by developer: 1
Blocking: |
Analyzed by developer: 0 |
-------------------------------------+-------------------------------------
Changes (by jyavenard):
* status: closed => reopened
* resolution: fixed =>
Comment:
Thank you for this...
However, are you sure the revert is complete?
The git comment itself is wrong, and the sha mentioned isn't the correct
one:
http://git.videolan.org/?p=ffmpeg.git;a=commit;h=5e50a5724bb00c44a98bd89f57c659a613f26ce2
The merged commit (from Libav) that removed lowres was
2bcbd98459915baefc15043d02f4a942ebcd33da
http://git.videolan.org/?p=ffmpeg.git;a=commit;h=2bcbd98459915baefc15043d02f4a942ebcd33da
it modified the files:
{{{
avconv.c
avplay.c
libavcodec/alpha/dsputil_alpha.c
libavcodec/arm/dsputil_init_arm.c
libavcodec/arm/dsputil_init_armv5te.c
libavcodec/arm/dsputil_init_armv6.c
libavcodec/arm/dsputil_init_neon.c
libavcodec/avcodec.h
libavcodec/dsputil.c
libavcodec/dsputil.h
libavcodec/dv.c
libavcodec/dvdec.c
libavcodec/error_resilience.c
libavcodec/flvdec.c
libavcodec/h261dec.c
libavcodec/h263dec.c
libavcodec/intrax8.c
libavcodec/jrevdct.c
libavcodec/libopenjpeg.c
libavcodec/mjpegbdec.c
libavcodec/mjpegdec.c
libavcodec/mpeg12.c
libavcodec/mpeg4videodec.c
libavcodec/mpegvideo.c
libavcodec/mpegvideo.h
libavcodec/msmpeg4.c
libavcodec/mxpegdec.c
libavcodec/options_table.h
libavcodec/ppc/dsputil_ppc.c
libavcodec/rv10.c
libavcodec/sp5xdec.c
libavcodec/utils.c
libavcodec/wmv2dec.c
libavcodec/x86/dsputil_mmx.c
}}}
Also, a following commits removed DSP code that were used for lowres
support:
d7458bc8c62ae1cb2ffc805b989fcddf4029dda6
http://git.videolan.org/?p=ffmpeg.git;a=commit;h=d7458bc8c62ae1cb2ffc805b989fcddf4029dda6
So the reversal made is only partial and only going to have an effect on
mpeg decoder...
BTW, I agree, there's no way four extra bit shifts would have caused a
drop of speed of 3%... Those stats could have only been made up...
--
Ticket URL: <https://ffmpeg.org/trac/ffmpeg/ticket/1351#comment:6>
FFmpeg <http://ffmpeg.org>
FFmpeg issue tracker
More information about the FFmpeg-trac
mailing list