March 2009 - FFmpeg-soc - lists.ffmpeg.org

[soc]: r4181 - libavfilter/doc/vfilters.texi
by victor 22 Mar '09

22 Mar '09

Author: victor Date: Sun Mar 22 17:24:02 2009 New Revision: 4181 Log: Forgot to escape the equal sign. Added a warning as penitence. Modified: libavfilter/doc/vfilters.texi Modified: libavfilter/doc/vfilters.texi ============================================================================== --- libavfilter/doc/vfilters.texi Sun Mar 22 17:15:05 2009 (r4180) +++ libavfilter/doc/vfilters.texi Sun Mar 22 17:24:02 2009 (r4181) @@ -156,12 +156,13 @@ Rotate video by a chosen amount in degre @section scale @example -./ffmpeg -i in.avi -vfilters "scale=200:100:sws_flags=bicubic" out.avi +./ffmpeg -i in.avi -vfilters "scale=200:100:sws_flags\=bicubic" out.avi @end example Scale to new width/height and/or convert pixel format. The two first parameter are the desired width and height. If the third parameter @var{sws_flags} is absent, then @var{bilinear} is assumed. +Be aware that the equal sign after @var{sws_flags} must be escaped, as in example. @section setpts

1 0

[soc]: r4180 - in libavfilter: README checkout.sh doc doc/vfilters.texi
by victor 22 Mar '09

22 Mar '09

Author: victor Date: Sun Mar 22 17:15:05 2009 New Revision: 4180 Log: Initial documentation of video filters. Added: libavfilter/doc/ libavfilter/doc/vfilters.texi Modified: libavfilter/README libavfilter/checkout.sh Modified: libavfilter/README ============================================================================== --- libavfilter/README Sun Mar 22 01:05:42 2009 (r4179) +++ libavfilter/README Sun Mar 22 17:15:05 2009 (r4180) @@ -68,22 +68,7 @@ of the filters. One of the interesting features of libavfilter is the ability for filters to have multiple inputs and multiple outputs. Unfortunately, ffplay cannot feed multiple video streams to the filter system, and cannot handle multiple output -streams. To illustrate the sorts of things that are possible, however, we can -use a complex filter graph. For example, the following graph: - -input --> split ---> fifo ----------------------> overlay --> output - | ^ - |------> fifo --> crop --> vflip --------| - -splits the stream into two streams, sends one stream through the crop filter -and the vflip filter before merging it back with the other stream by overlaying -it on top. You can use the following command to do this filtering - -./ffmpeg -i input_video.avi -s 240x320 -vfilters "[in]split [T1], fifo, [T2]overlay='0:240'[out]; [T1]fifo, crop=0:0:-1:240, vflip[T2]" out.avi - -where input_video.avi has a vertical resolution of 480 pixels, the result -should be that in output the top half of the video is mirroed onto the bottom -half. +streams. KNOWN ISSUES: - if for some reason the filter chain cannot be setup (due to an unknown Modified: libavfilter/checkout.sh ============================================================================== --- libavfilter/checkout.sh Sun Mar 22 01:05:42 2009 (r4179) +++ libavfilter/checkout.sh Sun Mar 22 17:15:05 2009 (r4180) @@ -8,3 +8,6 @@ for diff in $(ls $(pwd)/diffs/*.diff); d echo "copying files to libavfilter" find $(pwd) -maxdepth 1 -type f -not -name $(basename $0) -exec cp {} ffmpeg/libavfilter \; + +echo "copying libavfilter doc files" +cp -p $(pwd)/doc/* ffmpeg/doc Added: libavfilter/doc/vfilters.texi ============================================================================== --- /dev/null 00:00:00 1970 (empty, because file is newly added) +++ libavfilter/doc/vfilters.texi Sun Mar 22 17:15:05 2009 (r4180) @@ -0,0 +1,242 @@ +\input texinfo @c -*- texinfo -*- + +@settitle Video Filters Documentation +@titlepage +@sp 7 +@center @titlefont{Video Filters Documentation} +@sp 3 +@end titlepage + + +@chapter Introduction + +Libavfilter is the filtering API of FFmpeg. It is the substitute of the +now deprecated 'vhooks' and started as a Google Summer of Code project. +To get libavfilter you must download it into some directory of your choice by: +@example + svn checkout svn://svn.mplayerhq.hu/soc/libavfilter +@end example + +And then read @url{http://svn.mplayerhq.hu/soc/libavfilter/README?view=co} +to learn how to integrate it into ffmpeg and ffplay. + +The former video hook modules allowed a list of filters, that would run in the +order that they were specified on the ffmpeg command line. +Each filter had one input and one output. + +In libavfilter, it is possible for filters to have multiple inputs and +multiple outputs. +To illustrate the sorts of things that are possible, we can +use a complex filter graph. For example, the following one: + +@example +input --> split ---> fifo ----------------------> overlay --> output + | ^ + |------> fifo --> crop --> vflip --------| +@end example + +splits the stream into two streams, sends one stream through the crop filter +and the vflip filter before merging it back with the other stream by +overlaying it on top. You can use the following command to do this filtering + +@example +./ffmpeg -i in.avi -s 240x320 -vfilters "[in]split [T1], fifo, [T2]overlay='0:240'[out]; [T1]fifo, crop=0:0:-1:240, vflip[T2]" out.avi +@end example + +where input_video.avi has a vertical resolution of 480 pixels, the result +should be that in output the top half of the video is mirrored onto the bottom +half. + +Video filters are loaded using the @var{-vfilters} option to ffmpeg. The +filters in the same linear chain are separated by commas. In our example, +@var{split, fifo, overlay} are in one linear chain, and @var{fifo, crop, vflip} +are in another. +The points where the linear chains join are labeled by names enclosed in +square brackets. In our example, that is @var{[T1]} and @var{[T2]}. +The magic labels @var{[in]} and @var{[out]} are respectively the points +where video is input and output. +After the filter name, you may find the equal sign and a semicolon separated +parameter list. + +There are also the so called @var{source filters} that do not have a video input, +and we expect in future some @var{sink filters} that will not have video output. + +@chapter Available video filters + +When you configure your ffmpeg build, you can disable any of the existing video +filters. +The configure output will show the video filters included in your build. + +These are the currently available video filters: + +@section crop + +@example +./ffmpeg -i in.avi -vfilters "crop=0:0:-1:240" out.avi +@end example + +Crop the input video to x:y:width:height. +The -1 in width or height means that dimension in the input video. + +@section drawbox + +@example +./ffmpeg -i in.avi -vfilters "drawbox=10:20:200:60:red" out.avi +@end example + +Draw a box with x:y:width:height dimensions in a chosen color. + +@section fifo + +@example +./ffmpeg -i in.avi -vfilters "fifo" out.avi +@end example + +@section format + +@example +./ffmpeg -i in.avi -vfilters "format=yuv420p" out.avi +@end example + +Convert video to one of the specified filters formats (libavfilter will +try to pick one that is supported as an input to the next filter). + +@section fps + +@example +./ffmpeg -i in.avi -vfilters "fps=30000:1001" out.avi +@end example + +Modifies the framerate of the input video to a chosen one. + +@section hflip + +@example +./ffmpeg -i in.avi -vfilters "hflip" out.avi +@end example + +Flip the video horizontally. + +@section negate + +@example +./ffmpeg -i in.avi -vfilters "negate" out.avi +@end example + +Convert a video to its negative. + +@section noformat + +@example +./ffmpeg -i in.avi -vfilters "noformat=yuv420p" out.avi +@end example + +Force libavfilter not to send any of the specified pixel formats to the +next filter. + +@section overlay + +Overlay one video on top of another (see the example in the introduction +section). +Be aware that frames are taken +from each input video in timestamp order, hence, if their initial +timestamps differ, it is a a good idea to pass the two +inputs through a @var{setpts=PTS-STARTPTS} filter to have them +begin in the same zero timestamp, as it does the example for +the @var{movie} filter. + +@section rotate + +@example +./ffmpeg -i in.avi -vfilters "rotate=30" out.avi +@end example + +Rotate video by a chosen amount in degrees. By default, 45 degrees. + +@section scale + +@example +./ffmpeg -i in.avi -vfilters "scale=200:100:sws_flags=bicubic" out.avi +@end example + +Scale to new width/height and/or convert pixel format. +The two first parameter are the desired width and height. +If the third parameter @var{sws_flags} is absent, then @var{bilinear} is assumed. + +@section setpts + +@example +# Start counting PTS from zero +ffmpeg -i input.avi -vfilters setpts=PTS-STARTPTS output.avi + +# Fast motion +ffmpeg -i input.avi -vfilters setpts=0.5*PTS output.avi + +# Fixed rate 25 fps +ffmpeg -i input.avi -vfilters setpts=N*AVTB/25 output.avi + +# Fixed rate 25 fps with some jitter +ffmpeg -i input.avi -vfilters 'setpts=AVTB/25*(N+0.05*sin(N*2*PI/25))' output.avi +@end example + +Modifies the presentation timestamp (PTS) of the input video. + +@section slicify + +@example +./ffmpeg -i in.avi -vfilters "slicify=32" out.avi +@end example + +Pass on input video to next video filter as multiple slices. +The parameter is the slice height (16 if not specified). +Adding this in the beginning of filter chains should make filtering +faster due to the better use of the memory cache. + +@section split + +See the example in the introduction section. +The input video is passed on to two outputs. + +@section transpose + +@example +./ffmpeg -i in.avi -vfilters "transpose" out.avi +@end example + +Transpose (line => column) input video to next video filter. + +@section vflip + +@example +./ffmpeg -i in.avi -vfilters "vflip" out.avi +@end example + +Flip the video vertically. + +@section buffer + +This input filter is used by the client application to feed pictures to +the filter chain. See ffmpeg.c for an usage example. + +@section movie + +The parameters of the movie filter are +@example + seekpoint in microseconds : string format : string filename +@end example + +We can overlay a second movie on top of a main one as in this graph: + +@example + input -----------> deltapts0 --> overlay --> output + ^ + movie --> scale--> deltapts1 ------| +@end example + +To do that + +@example +ffmpeg -i in.avi -s 240x320 -vfilters "[in]setpts=PTS-STARTPTS, [T1]overlay=16:16[out]; movie=3200000:avi:in.avi, scale=180:144, setpts=PTS-STARTPTS[T1]" -y out.avi +@end example + +@bye

1 0

Interested in NEW Seeking API of GSoC09
by zhentan feng 20 Mar '09

20 Mar '09

hi, My name is zhentan feng, and I am a student from china. I have finished last year's soc project MXF muxer mentored by Baptiste. This summer, I plan to apply a soc project of ffmpeg again, and i am interested in the NEW Seeking API project. should I start from qualification task again or beginning to collect information for the new project? sincerely yours zhentan feng -- Best wishes~

1 0

[soc]: r4178 - aacenc/psymodel.c
by kostya 19 Mar '09

19 Mar '09

Author: kostya Date: Thu Mar 19 07:49:16 2009 New Revision: 4178 Log: Make frequency cutoff coefficient dependent on encoding parameters Modified: aacenc/psymodel.c Modified: aacenc/psymodel.c ============================================================================== --- aacenc/psymodel.c Wed Mar 18 22:19:58 2009 (r4177) +++ aacenc/psymodel.c Thu Mar 19 07:49:16 2009 (r4178) @@ -80,10 +80,17 @@ av_cold struct FFPsyPreprocessContext* f { FFPsyPreprocessContext *ctx; int i; + float cutoff_coeff; ctx = av_mallocz(sizeof(FFPsyPreprocessContext)); ctx->avctx = avctx; + + if(avctx->flags & CODEC_FLAG_QSCALE) + cutoff_coeff = 1.0f / av_clip(1 + avctx->global_quality / FF_QUALITY_SCALE, 1, 8); + else + cutoff_coeff = avctx->bit_rate / (4.0f * avctx->sample_rate * avctx->channels); + ctx->fcoeffs = ff_iir_filter_init_coeffs(FF_FILTER_TYPE_BUTTERWORTH, FF_FILTER_MODE_LOWPASS, - FILT_ORDER, 0.25, 0.0, 0.0); + FILT_ORDER, cutoff_coeff, 0.0, 0.0); if(ctx->fcoeffs){ ctx->fstate = av_mallocz(sizeof(ctx->fstate[0]) * avctx->channels); for(i = 0; i < avctx->channels; i++)

1 0

[soc]: r4177 - wmapro/wma3dec.c
by faust3 18 Mar '09

18 Mar '09

Author: faust3 Date: Wed Mar 18 22:19:58 2009 New Revision: 4177 Log: do not recalculate the minimum subframe length in wma_decode_tilehdr Modified: wmapro/wma3dec.c Modified: wmapro/wma3dec.c ============================================================================== --- wmapro/wma3dec.c Wed Mar 18 22:05:49 2009 (r4176) +++ wmapro/wma3dec.c Wed Mar 18 22:19:58 2009 (r4177) @@ -453,7 +453,7 @@ static int wma_decode_tilehdr(WMA3Decode int min_channel_len = s->samples_per_frame; int read_channel_mask = 1; int channels_for_cur_subframe = 0; - int subframe_len = s->samples_per_frame / s->max_num_subframes; + int subframe_len; if(fixed_channel_layout){ read_channel_mask = 0; @@ -472,7 +472,7 @@ static int wma_decode_tilehdr(WMA3Decode } if(channels_for_cur_subframe == 1 || - subframe_len * channels_for_cur_subframe == missing_samples) + s->min_samples_per_subframe * channels_for_cur_subframe == missing_samples) read_channel_mask = 0; } @@ -488,9 +488,8 @@ static int wma_decode_tilehdr(WMA3Decode } } - /** if the frames are not evenly split, get the next subframe - length from the bitstream */ - if(subframe_len * channels_for_cur_subframe != missing_samples){ + /** if we have the choice get next subframe length from the bitstream */ + if(s->min_samples_per_subframe * channels_for_cur_subframe != missing_samples){ int log2_subframe_len = 0; /* 1 bit indicates if the subframe length is zero */ if(subframe_len_zero_bit){ @@ -517,7 +516,8 @@ static int wma_decode_tilehdr(WMA3Decode "broken frame: subframe_len %i\n", subframe_len); return -1; } - } + }else + subframe_len = s->min_samples_per_subframe; for(c=0; c<s->num_channels;c++){ WMA3ChannelCtx* chan = &s->channel[c];

1 0

[soc]: r4176 - in wmapro: wma3.h wma3dec.c
by faust3 18 Mar '09

18 Mar '09

Author: faust3 Date: Wed Mar 18 22:05:49 2009 New Revision: 4176 Log: simplified wma_decode_tilehdr a bit Modified: wmapro/wma3.h wmapro/wma3dec.c Modified: wmapro/wma3.h ============================================================================== --- wmapro/wma3.h Tue Mar 17 18:35:02 2009 (r4175) +++ wmapro/wma3.h Wed Mar 18 22:05:49 2009 (r4176) @@ -137,7 +137,6 @@ typedef struct WMA3DecodeContext { int16_t* samples; //< current samplebuffer pointer int16_t* samples_end; //< maximum samplebuffer pointer uint8_t drc_gain; //< gain for the DRC tool - uint8_t no_tiling; //< frames contain subframes int skip_frame; //< skip output step int parsed_all_subframes; //< all subframes decoded? Modified: wmapro/wma3dec.c ============================================================================== --- wmapro/wma3dec.c Tue Mar 17 18:35:02 2009 (r4175) +++ wmapro/wma3dec.c Wed Mar 18 22:05:49 2009 (r4176) @@ -433,8 +433,10 @@ static int wma_decode_tilehdr(WMA3Decode } }else{ /** subframe length and number of subframes is not constant */ int subframe_len_bits = 0; /** bits needed for the subframe length */ - int subframe_len_zero_bit = 0; /** how many of the length bits indicate - if the subframe is zero */ + int subframe_len_zero_bit = 0; /** first bit indicates if length is zero */ + int fixed_channel_layout; /** all channels have the same subframe layout */ + + fixed_channel_layout = get_bits1(&s->getbit); /** calculate subframe len bits */ if(s->lossless) @@ -447,63 +449,48 @@ static int wma_decode_tilehdr(WMA3Decode /** loop until the frame data is split between the subframes */ while(missing_samples > 0){ - int64_t tileinfo = -1; + unsigned int channel_mask = 0; int min_channel_len = s->samples_per_frame; - int num_subframes_per_channel = 0; - int num_channels = 0; + int read_channel_mask = 1; + int channels_for_cur_subframe = 0; int subframe_len = s->samples_per_frame / s->max_num_subframes; - /** find channel with the smallest overall length */ - for(c=0;c<s->num_channels;c++){ - if(min_channel_len > s->channel[c].channel_len) - min_channel_len = s->channel[c].channel_len; - } - - /** check if this is the start of a new frame */ - if(missing_samples == s->num_channels * s->samples_per_frame){ - s->no_tiling = get_bits1(&s->getbit); - } - - if(s->no_tiling){ - num_subframes_per_channel = 1; - num_channels = s->num_channels; + if(fixed_channel_layout){ + read_channel_mask = 0; + channels_for_cur_subframe = s->num_channels; + min_channel_len = s->channel[0].channel_len; }else{ - /** count how many channels have the minimum length */ + /** find channels with the smallest overall length */ for(c=0;c<s->num_channels;c++){ - if(min_channel_len == s->channel[c].channel_len){ - ++num_channels; + if(s->channel[c].channel_len <= min_channel_len){ + if(s->channel[c].channel_len < min_channel_len){ + channels_for_cur_subframe = 0; + min_channel_len = s->channel[c].channel_len; + } + ++channels_for_cur_subframe; } } - if(num_channels <= 1) - num_subframes_per_channel = 1; - } - /** maximum number of subframes, evenly split */ - if(subframe_len == missing_samples / num_channels){ - num_subframes_per_channel = 1; + if(channels_for_cur_subframe == 1 || + subframe_len * channels_for_cur_subframe == missing_samples) + read_channel_mask = 0; } - /** if there might be multiple subframes per channel */ - if(num_subframes_per_channel != 1){ - int total_num_bits = num_channels; - tileinfo = 0; - /** For every channel with the minimum length, 1 bit - is transmitted that informs us if the channel - contains a subframe with the next subframe_len. */ - while(total_num_bits){ - int num_bits = total_num_bits; - if(num_bits > 32) - num_bits = 32; - tileinfo |= get_bits_long(&s->getbit,num_bits); - total_num_bits -= num_bits; - num_bits = total_num_bits; - tileinfo <<= (num_bits > 32)? 32 : num_bits; + /** For every channel with the minimum length, 1 bit + might be transmitted that informs us if the channel + contains a subframe with the next subframe_len. */ + if(read_channel_mask){ + channel_mask = get_bits(&s->getbit,channels_for_cur_subframe); + if(!channel_mask){ + av_log(s->avctx, AV_LOG_ERROR, + "broken frame: zero frames for subframe_len\n"); + return -1; } } /** if the frames are not evenly split, get the next subframe length from the bitstream */ - if(subframe_len != missing_samples / num_channels){ + if(subframe_len * channels_for_cur_subframe != missing_samples){ int log2_subframe_len = 0; /* 1 bit indicates if the subframe length is zero */ if(subframe_len_zero_bit){ @@ -522,28 +509,23 @@ static int wma_decode_tilehdr(WMA3Decode }else subframe_len = s->samples_per_frame / (1 << log2_subframe_len); - } - /** sanity check the length */ - if(subframe_len < s->min_samples_per_subframe - || subframe_len > s->samples_per_frame){ - av_log(s->avctx, AV_LOG_ERROR, + /** sanity check the length */ + if(subframe_len < s->min_samples_per_subframe + || subframe_len > s->samples_per_frame){ + av_log(s->avctx, AV_LOG_ERROR, "broken frame: subframe_len %i\n", subframe_len); - return -1; + return -1; + } } + for(c=0; c<s->num_channels;c++){ WMA3ChannelCtx* chan = &s->channel[c]; - if(chan->num_subframes > 32){ - av_log(s->avctx, AV_LOG_ERROR, - "broken frame: num subframes %i\n", - chan->num_subframes); - return -1; - } /** add subframes to the individual channels */ if(min_channel_len == chan->channel_len){ - --num_channels; - if(tileinfo & (1<<num_channels)){ + --channels_for_cur_subframe; + if(!read_channel_mask || channel_mask & (1<<channels_for_cur_subframe)){ if(chan->num_subframes > 31){ av_log(s->avctx, AV_LOG_ERROR, "broken frame: num subframes > 31\n");

1 0

[soc]: r4172 - aacenc/aacenc.c
by kostya 17 Mar '09

17 Mar '09

Author: kostya Date: Tue Mar 17 18:24:05 2009 New Revision: 4172 Log: Fill all elements in ff_aac_pow2sf_tab[]; its size has been increased Modified: aacenc/aacenc.c Modified: aacenc/aacenc.c ============================================================================== --- aacenc/aacenc.c Tue Mar 17 18:19:21 2009 (r4171) +++ aacenc/aacenc.c Tue Mar 17 18:24:05 2009 (r4172) @@ -197,7 +197,7 @@ static av_cold int aac_encode_init(AVCod s->coder = &ff_aac_coders[0]; #if !CONFIG_HARDCODED_TABLES - for (i = 0; i < 316; i++) + for (i = 0; i < 428; i++) ff_aac_pow2sf_tab[i] = pow(2, (i - 200)/4.); #endif /* CONFIG_HARDCODED_TABLES */

3 2

[soc]: r4175 - in aacenc: aaccoder.c aacenc.c
by kostya 17 Mar '09

17 Mar '09

Author: kostya Date: Tue Mar 17 18:35:02 2009 New Revision: 4175 Log: Spread TODOs between corresponding files Modified: aacenc/aaccoder.c aacenc/aacenc.c Modified: aacenc/aaccoder.c ============================================================================== --- aacenc/aaccoder.c Tue Mar 17 18:32:36 2009 (r4174) +++ aacenc/aaccoder.c Tue Mar 17 18:35:02 2009 (r4175) @@ -24,6 +24,12 @@ * AAC coefficients encoder */ +/*********************************** + * TODOs: + * speedup quantizer selection + * add sane pulse detection + ***********************************/ + #include "avcodec.h" #include "bitstream.h" #include "aac.h" Modified: aacenc/aacenc.c ============================================================================== --- aacenc/aacenc.c Tue Mar 17 18:32:36 2009 (r4174) +++ aacenc/aacenc.c Tue Mar 17 18:35:02 2009 (r4175) @@ -26,8 +26,6 @@ /*********************************** * TODOs: - * speedup quantizer selection - * add sane pulse detection * add temporal noise shaping ***********************************/

1 0

[soc]: r4174 - aacenc/aaccoder.c
by kostya 17 Mar '09

17 Mar '09

Author: kostya Date: Tue Mar 17 18:32:36 2009 New Revision: 4174 Log: Introduce a new function for choosing codebooks with emphasis on minimizing distortion with intended use in non-optimal but fast coding. Modified: aacenc/aaccoder.c Modified: aacenc/aaccoder.c ============================================================================== --- aacenc/aaccoder.c Tue Mar 17 18:26:55 2009 (r4173) +++ aacenc/aaccoder.c Tue Mar 17 18:32:36 2009 (r4174) @@ -562,6 +562,13 @@ static void encode_window_bands_info(AAC } } +static void encode_window_bands_info_fixed(AACEncContext *s, SingleChannelElement *sce, + int win, int group_len, const float lambda) +{ + encode_window_bands_info(s, sce, win, group_len, 1.0f); +} + + typedef struct TrellisPath { float cost; int prev;

1 0

[soc]: r4173 - in aacenc: aaccoder.c aacenc.c aacenc.h
by kostya 17 Mar '09

17 Mar '09

Author: kostya Date: Tue Mar 17 18:26:55 2009 New Revision: 4173 Log: Pass lambda as parameter to encode_window_bands_info() as well Modified: aacenc/aaccoder.c aacenc/aacenc.c aacenc/aacenc.h Modified: aacenc/aaccoder.c ============================================================================== --- aacenc/aaccoder.c Tue Mar 17 18:24:05 2009 (r4172) +++ aacenc/aaccoder.c Tue Mar 17 18:26:55 2009 (r4173) @@ -467,7 +467,7 @@ typedef struct BandCodingPath { * Encode band info for single window group bands. */ static void encode_window_bands_info(AACEncContext *s, SingleChannelElement *sce, - int win, int group_len) + int win, int group_len, const float lambda) { BandCodingPath path[120][12]; int w, swb, cb, start, start2, size; @@ -502,7 +502,7 @@ static void encode_window_bands_info(AAC FFPsyBand *band = &s->psy.psy_bands[s->cur_channel*PSY_MAX_BANDS+(win+w)*16+swb]; rd += quantize_band_cost(sce->coeffs + start + w*128, size, sce->sf_idx[(win+w)*16+swb], cb, - s->lambda / band->threshold, INFINITY, NULL); + lambda / band->threshold, INFINITY, NULL); } if( run_value_bits[sce->ics.num_windows == 8][path[swb][cb].run] != run_value_bits[sce->ics.num_windows == 8][path[swb][cb].run+1]) Modified: aacenc/aacenc.c ============================================================================== --- aacenc/aacenc.c Tue Mar 17 18:24:05 2009 (r4172) +++ aacenc/aacenc.c Tue Mar 17 18:26:55 2009 (r4173) @@ -362,7 +362,7 @@ static void encode_band_info(AACEncConte int w; for(w = 0; w < sce->ics.num_windows; w += sce->ics.group_len[w]){ - s->coder->encode_window_bands_info(s, sce, w, sce->ics.group_len[w]); + s->coder->encode_window_bands_info(s, sce, w, sce->ics.group_len[w], s->lambda); } } Modified: aacenc/aacenc.h ============================================================================== --- aacenc/aacenc.h Tue Mar 17 18:24:05 2009 (r4172) +++ aacenc/aacenc.h Tue Mar 17 18:26:55 2009 (r4173) @@ -36,7 +36,7 @@ typedef struct AACCoefficientsEncoder{ void (*search_for_quantizers)(AVCodecContext *avctx, struct AACEncContext *s, SingleChannelElement *sce, const float lambda); void (*encode_window_bands_info)(struct AACEncContext *s, SingleChannelElement *sce, - int win, int group_len); + int win, int group_len, const float lambda); void (*quantize_and_encode_band)(PutBitContext *pb, const float *in, int size, int scale_idx, int cb, const float lambda); void (*search_for_ms)(struct AACEncContext *s, ChannelElement *cpe, const float lambda);

1 0