December 2009 - ffmpeg-devel - lists.ffmpeg.org

[PATCH] Fix vflip alpha plane settings
by stefano.sabatini-lala＠poste.it 08 Dec '09

08 Dec '09

Hi, this fixes: "format=yuva420p, vflip" +10l to me. Regards. -- FFmpeg = Freak & Frenzy Majestic Peaceful Elected Game

2 2

BSAC Decoder Support ?
by amar.ajeet＠gmail.com 08 Dec '09

08 Dec '09

Hi All, Do any one have Patch for BSAC decoder . Please give me details how can we implement the BSAC decoder . Thanks Rishi

2 1

[PATCH] Make VP3/Theora Decoder Much Faster
by mike＠multimedia.cx 07 Dec '09

07 Dec '09

$subject For further explanation, including specific performance numbers, see: http://multimedia.cx/eggs/one-gluttonous-if-statement/ I know that I "own" the Theora decoder so it's my prerogative to commit this change. But I'm still hoping for input before committing. Cosmetically, I'm not happy with the naming of fast_fragment_list[]. But that's the best name I could come up with whilst in the momentum of coding. -- -Mike Melanson

5 15

Deinterlacing
by pritula＠yandex-team.ru 07 Dec '09

07 Dec '09

Hi, Fellows. I noticed, that FFmpeg doesn't watch on AVFrame field interlaced_frame and without option -deinterlace never does deinterlacing, and with option -deinterlace always does deinterlacing. May be I mistaken, but it seems to me, that FFmpeg should be able to deinterlace only interlaced frames. Something like -deinterlace-if-needed option. If this makes sense, I will post patch with new option soon. What do you think? -- Mikhail Pritula

3 4

Problem with initial dts in h264
by stebbins＠jetheaddev.com 07 Dec '09

07 Dec '09

A recent change has caused the initial dts in h264 streams returned by av_read_frame to be incorrect. The commit was r20599. The root cause of the problem comes from moving the assignment of s->avctx->pix_fmt from h264.c:decode_init() to decode_slice_header(). This movement has the side effect of making has_codec_parameters() in libavformat/utils.c fail during av_find_stream_info(). This causes try_decode_frame() to be called which ultimately sets st->codec->has_b_frames = 2. This the causes compute_pkt_fields() in libavformat/utils.c to start buffering up multiple pts values in st->pts_buffer. All this happens during the initial call to av_find_stream_info. Then later, when the application calls av_read_frame, the pts values that were buffered during the previous av_find_stream_info are used to set the dts value of the returned packet. This is especially obnoxious when the reader does a seek before av_read_frame. The reason this worked prior to r20599 is that st->codec->has_b_frames would only be set to 0 during av_find_stream_info, so only the first entry of st->pts_buffer would ever be filled in. And this first entry is always replaced on the next iteration.

4 7

[PATCH] AU : demuxed packet size should be sample size aligned
by jmenon86＠gmail.com 07 Dec '09

07 Dec '09

Hi, As in subject. Fixes issue 1593. -- Jai Menon

2 4

[PATCH] MMX2 scaler: Determine "funnyCode" size at runtime.
by ramiro.polla＠gmail.com 07 Dec '09

07 Dec '09

Hi, $subj again. Ramiro Polla

2 8

[PATCH] internal error argb -> rgba converter
by stefano.sabatini-lala＠poste.it 07 Dec '09

07 Dec '09

Hi, stefano at geppetto ~/s/l/ffmpeg> ffplay -loglevel info in.avi -vfilters "format=argb, format=rgba" [...] [format @ 0x94bc1a0]auto-inserting filter 'scale' [format @ 0x94ca650]auto-inserting filter 'scale' [ffplay_output @ 0x94bbb10]auto-inserting filter 'scale' [swscaler @ 0x94cb610]No accelerated colorspace conversion found. [scale @ 0x94cae70]w:464 h:352 fmt:argb [scale @ 0x94cb0d0]w:464 h:352 fmt:rgba [scale @ 0x94cb5b0]w:464 h:352 fmt:yuv420p [swscaler @ 0x94dacb0]internal error argb -> rgba converter 1.41 A-V: 0.005 aq= 246KB vq= 1037KB sq= 0B Note that the reverse conversion (rgba -> argb) works fine. Can someone have a look at that? Regards. -- FFmpeg = Fiendish and Fostering Mean Powered Enhancing Glue

3 4

[PATCH] Start using intermediate buffers at index 0.
by ramiro.polla＠gmail.com 06 Dec '09

06 Dec '09

Hi, These index variables are incremented before each use, so they should be initialized to -1. Ramiro Polla

2 2

[PATCH] Make slicify honour the slice direction
by stefano.sabatini-lala＠poste.it 06 Dec '09

06 Dec '09

Hi, as in subject. Without this some combination with the pad filter may issue weird behavior. Regards. -- FFmpeg = Fast Furious Mere Peaceful Eccentric Guru

2 7